fix(db): Remove redundant migration and fix pgvector schema handling (#11822)

### Changes 🏗️

This PR includes two database migration fixes:

#### 1. Remove redundant Supabase extensions migration

Removes the `20260112173500_add_supabase_extensions_to_platform_schema`
migration which was attempting to manage Supabase-provided extensions
and schemas.

**What was removed:**
- Migration that created extensions (pgcrypto, uuid-ossp,
pg_stat_statements, pg_net, pgjwt, pg_graphql, pgsodium, supabase_vault)
- Schema creation for these extensions

**Why it was removed:**
- These extensions and schemas are pre-installed and managed by Supabase
automatically
- The migration was redundant and could cause schema drift warnings
- Attempting to manage Supabase-owned resources in our migrations is an
anti-pattern

#### 2. Fix pgvector extension schema handling

Improves the `20260109181714_add_docs_embedding` migration to handle
cases where pgvector exists in the wrong schema.

**Problem:**
- If pgvector was previously installed in `public` schema, `CREATE
EXTENSION IF NOT EXISTS` would succeed but not actually install it in
the `platform` schema
- This causes `type "vector" does not exist` errors because the type
isn't in the search_path

**Solution:**
- Detect if vector extension exists in a different schema than the
current one
- Drop it with CASCADE and reinstall in the correct schema (platform)
- Use dynamic SQL with `EXECUTE format()` to explicitly specify the
target schema
- Split exception handling: catch errors during removal, but let
installation fail naturally with clear PostgreSQL errors

**Impact:**
- No functional changes - Supabase continues to provide extensions as
before
- pgvector now correctly installs in the platform schema
- Cleaner migration history
- Prevents schema-related errors

### Checklist 📋

#### For code changes:
- [x] I have clearly listed my changes in the PR description
- [x] I have made a test plan
- [x] I have tested my changes according to the test plan:
  - [x] Verified migrations run successfully without the redundant file
  - [x] Confirmed Supabase extensions are still available
  - [x] Tested pgvector migration handles wrong-schema scenario
  - [x] No schema drift warnings

#### For configuration changes:
- [x] .env.default is updated or already compatible with my changes
- [x] docker-compose.yml is updated or already compatible with my
changes
- [x] I have included a list of my configuration changes in the PR
description (under **Changes**)
  - N/A - No configuration changes required

.github/workflows/platform-frontend-ci.yml

@@ -128,7 +128,7 @@ jobs:
           token: ${{ secrets.GITHUB_TOKEN }}
           exitOnceUploaded: true
 
-  test:
+  e2e_test:
     runs-on: big-boi
     needs: setup
     strategy:
@@ -258,3 +258,39 @@ jobs:
       - name: Print Final Docker Compose logs
         if: always()
         run: docker compose -f ../docker-compose.yml logs
+
+  integration_test:
+    runs-on: ubuntu-latest
+    needs: setup
+
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@v4
+        with:
+          submodules: recursive
+
+      - name: Set up Node.js
+        uses: actions/setup-node@v4
+        with:
+          node-version: "22.18.0"
+
+      - name: Enable corepack
+        run: corepack enable
+
+      - name: Restore dependencies cache
+        uses: actions/cache@v4
+        with:
+          path: ~/.pnpm-store
+          key: ${{ needs.setup.outputs.cache-key }}
+          restore-keys: |
+            ${{ runner.os }}-pnpm-${{ hashFiles('autogpt_platform/frontend/pnpm-lock.yaml') }}
+            ${{ runner.os }}-pnpm-
+
+      - name: Install dependencies
+        run: pnpm install --frozen-lockfile
+
+      - name: Generate API client
+        run: pnpm generate:api
+
+      - name: Run Integration Tests
+        run: pnpm test:unit

AGENTS.md

@@ -16,6 +16,32 @@ See `docs/content/platform/getting-started.md` for setup instructions.
 - Format Python code with `poetry run format`.
 - Format frontend code using `pnpm format`.
 
+
+## Frontend guidelines:
+
+See `/frontend/CONTRIBUTING.md` for complete patterns. Quick reference:
+
+1. **Pages**: Create in `src/app/(platform)/feature-name/page.tsx`
+   - Add `usePageName.ts` hook for logic
+   - Put sub-components in local `components/` folder
+2. **Components**: Structure as `ComponentName/ComponentName.tsx` + `useComponentName.ts` + `helpers.ts`
+   - Use design system components from `src/components/` (atoms, molecules, organisms)
+   - Never use `src/components/__legacy__/*`
+3. **Data fetching**: Use generated API hooks from `@/app/api/__generated__/endpoints/`
+   - Regenerate with `pnpm generate:api`
+   - Pattern: `use{Method}{Version}{OperationName}`
+4. **Styling**: Tailwind CSS only, use design tokens, Phosphor Icons only
+5. **Testing**: Add Storybook stories for new components, Playwright for E2E
+6. **Code conventions**: Function declarations (not arrow functions) for components/handlers
+- Component props should be `interface Props { ... }` (not exported) unless the interface needs to be used outside the component
+- Separate render logic from business logic (component.tsx + useComponent.ts + helpers.ts)
+- Colocate state when possible and avoid creating large components, use sub-components ( local `/components` folder next to the parent component ) when sensible
+- Avoid large hooks, abstract logic into `helpers.ts` files when sensible
+- Use function declarations for components, arrow functions only for callbacks
+- No barrel files or `index.ts` re-exports
+- Do not use `useCallback` or `useMemo` unless strictly needed
+- Avoid comments at all times unless the code is very complex
+
 ## Testing
 
 - Backend: `poetry run test` (runs pytest with a docker based postgres + prisma).

autogpt_platform/CLAUDE.md

@@ -201,7 +201,7 @@ If you get any pushback or hit complex block conditions check the new_blocks gui
 3. Write tests alongside the route file
 4. Run `poetry run test` to verify
 
-**Frontend feature development:**
+### Frontend guidelines:
 
 See `/frontend/CONTRIBUTING.md` for complete patterns. Quick reference:
 
@@ -217,6 +217,14 @@ See `/frontend/CONTRIBUTING.md` for complete patterns. Quick reference:
 4. **Styling**: Tailwind CSS only, use design tokens, Phosphor Icons only
 5. **Testing**: Add Storybook stories for new components, Playwright for E2E
 6. **Code conventions**: Function declarations (not arrow functions) for components/handlers
+- Component props should be `interface Props { ... }` (not exported) unless the interface needs to be used outside the component
+- Separate render logic from business logic (component.tsx + useComponent.ts + helpers.ts)
+- Colocate state when possible and avoid creating large components, use sub-components ( local `/components` folder next to the parent component ) when sensible
+- Avoid large hooks, abstract logic into `helpers.ts` files when sensible
+- Use function declarations for components, arrow functions only for callbacks
+- No barrel files or `index.ts` re-exports
+- Do not use `useCallback` or `useMemo` unless strictly needed
+- Avoid comments at all times unless the code is very complex
 
 ### Security Implementation
 

autogpt_platform/backend/backend/api/features/builder/db.py

@@ -1,17 +1,16 @@
 import logging
 from dataclasses import dataclass
+from datetime import datetime, timedelta, timezone
 from difflib import SequenceMatcher
 from typing import Sequence
 
 import prisma
-from prisma.enums import ContentType
 
 import backend.api.features.library.db as library_db
 import backend.api.features.library.model as library_model
 import backend.api.features.store.db as store_db
 import backend.api.features.store.model as store_model
 import backend.data.block
-from backend.api.features.store.hybrid_search import unified_hybrid_search
 from backend.blocks import load_all_blocks
 from backend.blocks.llm import LlmModel
 from backend.data.block import AnyBlockSchema, BlockCategory, BlockInfo, BlockSchema
@@ -38,14 +37,6 @@ MAX_LIBRARY_AGENT_RESULTS = 100
 MAX_MARKETPLACE_AGENT_RESULTS = 100
 MIN_SCORE_FOR_FILTERED_RESULTS = 10.0
 
-# Boost blocks over marketplace agents in search results
-BLOCK_SCORE_BOOST = 50.0
-
-# Block IDs to exclude from search results
-EXCLUDED_BLOCK_IDS = frozenset({
-    "e189baac-8c20-45a1-94a7-55177ea42565",  # AgentExecutorBlock
-})
-
 SearchResultItem = BlockInfo | library_model.LibraryAgent | store_model.StoreAgent
 
 
@@ -259,25 +250,14 @@ async def _build_cached_search_results(
         "my_agents": 0,
     }
 
-    # Use hybrid search when query is present, otherwise list all blocks
-    if (include_blocks or include_integrations) and normalized_query:
-        block_results, block_total, integration_total = await _hybrid_search_blocks(
-            query=search_query,
-            include_blocks=include_blocks,
-            include_integrations=include_integrations,
-        )
-        scored_items.extend(block_results)
-        total_items["blocks"] = block_total
-        total_items["integrations"] = integration_total
-    elif include_blocks or include_integrations:
-        # No query - list all blocks using in-memory approach
-        block_results, block_total, integration_total = _collect_block_results(
-            include_blocks=include_blocks,
-            include_integrations=include_integrations,
-        )
-        scored_items.extend(block_results)
-        total_items["blocks"] = block_total
-        total_items["integrations"] = integration_total
+    block_results, block_total, integration_total = _collect_block_results(
+        normalized_query=normalized_query,
+        include_blocks=include_blocks,
+        include_integrations=include_integrations,
+    )
+    scored_items.extend(block_results)
+    total_items["blocks"] = block_total
+    total_items["integrations"] = integration_total
 
     if include_library_agents:
         library_response = await library_db.list_library_agents(
@@ -322,14 +302,10 @@ async def _build_cached_search_results(
 
 def _collect_block_results(
     *,
+    normalized_query: str,
     include_blocks: bool,
     include_integrations: bool,
 ) -> tuple[list[_ScoredItem], int, int]:
-    """
-    Collect all blocks for listing (no search query).
-
-    All blocks get BLOCK_SCORE_BOOST to prioritize them over marketplace agents.
-    """
     results: list[_ScoredItem] = []
     block_count = 0
     integration_count = 0
@@ -342,10 +318,6 @@ def _collect_block_results(
         if block.disabled:
             continue
 
-        # Skip excluded blocks
-        if block.id in EXCLUDED_BLOCK_IDS:
-            continue
-
         block_info = block.get_info()
         credentials = list(block.input_schema.get_credentials_fields().values())
         is_integration = len(credentials) > 0
@@ -355,6 +327,10 @@ def _collect_block_results(
         if not is_integration and not include_blocks:
             continue
 
+        score = _score_block(block, block_info, normalized_query)
+        if not _should_include_item(score, normalized_query):
+            continue
+
         filter_type: FilterType = "integrations" if is_integration else "blocks"
         if is_integration:
             integration_count += 1
@@ -365,116 +341,8 @@ def _collect_block_results(
             _ScoredItem(
                 item=block_info,
                 filter_type=filter_type,
-                score=BLOCK_SCORE_BOOST,
-                sort_key=block_info.name.lower(),
-            )
-        )
-
-    return results, block_count, integration_count
-
-
-async def _hybrid_search_blocks(
-    *,
-    query: str,
-    include_blocks: bool,
-    include_integrations: bool,
-) -> tuple[list[_ScoredItem], int, int]:
-    """
-    Search blocks using hybrid search with builder-specific filtering.
-
-    Uses unified_hybrid_search for semantic + lexical search, then applies
-    post-filtering for block/integration types and LLM model bonus scoring.
-
-    Args:
-        query: The search query string
-        include_blocks: Whether to include regular blocks
-        include_integrations: Whether to include integration blocks
-
-    Returns:
-        Tuple of (scored_items, block_count, integration_count)
-    """
-    results: list[_ScoredItem] = []
-    block_count = 0
-    integration_count = 0
-
-    if not include_blocks and not include_integrations:
-        return results, block_count, integration_count
-
-    normalized_query = query.strip().lower()
-
-    # Fetch more results to account for post-filtering
-    search_results, _ = await unified_hybrid_search(
-        query=query,
-        content_types=[ContentType.BLOCK],
-        page=1,
-        page_size=150,
-        min_score=0.10,
-    )
-
-    # Load all blocks for getting BlockInfo
-    all_blocks = load_all_blocks()
-
-    for result in search_results:
-        block_id = result["content_id"]
-
-        # Skip excluded blocks
-        if block_id in EXCLUDED_BLOCK_IDS:
-            continue
-
-        metadata = result.get("metadata", {})
-        hybrid_score = result.get("relevance", 0.0)
-
-        # Get the actual block class
-        if block_id not in all_blocks:
-            continue
-
-        block_cls = all_blocks[block_id]
-        block: AnyBlockSchema = block_cls()
-
-        if block.disabled:
-            continue
-
-        # Check block/integration filter using metadata
-        is_integration = metadata.get("is_integration", False)
-
-        if is_integration and not include_integrations:
-            continue
-        if not is_integration and not include_blocks:
-            continue
-
-        # Get block info
-        block_info = block.get_info()
-
-        # Calculate final score: scale hybrid score and add builder-specific bonuses
-        # Hybrid scores are 0-1, builder scores were 0-200+
-        # Add BLOCK_SCORE_BOOST to prioritize blocks over marketplace agents
-        final_score = hybrid_score * 100 + BLOCK_SCORE_BOOST
-
-        # Add LLM model match bonus
-        has_llm_field = metadata.get("has_llm_model_field", False)
-        if has_llm_field and _matches_llm_model(block.input_schema, normalized_query):
-            final_score += 20
-
-        # Add exact/prefix match bonus for deterministic tie-breaking
-        name = block_info.name.lower()
-        if name == normalized_query:
-            final_score += 30
-        elif name.startswith(normalized_query):
-            final_score += 15
-
-        # Track counts
-        filter_type: FilterType = "integrations" if is_integration else "blocks"
-        if is_integration:
-            integration_count += 1
-        else:
-            block_count += 1
-
-        results.append(
-            _ScoredItem(
-                item=block_info,
-                filter_type=filter_type,
-                score=final_score,
-                sort_key=name,
+                score=score,
+                sort_key=_get_item_name(block_info),
             )
         )
 
@@ -634,6 +502,38 @@ def _matches_llm_model(schema_cls: type[BlockSchema], query: str) -> bool:
     return False
 
 
+def _score_block(
+    block: AnyBlockSchema,
+    block_info: BlockInfo,
+    normalized_query: str,
+) -> float:
+    if not normalized_query:
+        return 0.0
+
+    name = block_info.name.lower()
+    description = block_info.description.lower()
+    score = _score_primary_fields(name, description, normalized_query)
+
+    category_text = " ".join(
+        category.get("category", "").lower() for category in block_info.categories
+    )
+    score += _score_additional_field(category_text, normalized_query, 12, 6)
+
+    credentials_info = block.input_schema.get_credentials_fields_info().values()
+    provider_names = [
+        provider.value.lower()
+        for info in credentials_info
+        for provider in info.provider
+    ]
+    provider_text = " ".join(provider_names)
+    score += _score_additional_field(provider_text, normalized_query, 15, 6)
+
+    if _matches_llm_model(block.input_schema, normalized_query):
+        score += 20
+
+    return score
+
+
 def _score_library_agent(
     agent: library_model.LibraryAgent,
     normalized_query: str,
@@ -740,15 +640,26 @@ def _get_all_providers() -> dict[ProviderName, Provider]:
     return providers
 
 
-@cached(ttl_seconds=3600, shared_cache=True)
+@cached(ttl_seconds=3600)
 async def get_suggested_blocks(count: int = 5) -> list[BlockInfo]:
-    # Query the materialized view for execution counts per block
-    # The view aggregates executions from the last 14 days and is refreshed hourly
+    suggested_blocks = []
+    # Sum the number of executions for each block type
+    # Prisma cannot group by nested relations, so we do a raw query
+    # Calculate the cutoff timestamp
+    timestamp_threshold = datetime.now(timezone.utc) - timedelta(days=30)
+
     results = await query_raw_with_schema(
         """
-        SELECT block_id, execution_count
-        FROM {schema_prefix}"mv_suggested_blocks";
-        """
+        SELECT
+            agent_node."agentBlockId" AS block_id,
+            COUNT(execution.id) AS execution_count
+        FROM {schema_prefix}"AgentNodeExecution" execution
+        JOIN {schema_prefix}"AgentNode" agent_node ON execution."agentNodeId" = agent_node.id
+        WHERE execution."endedTime" >= $1::timestamp
+        GROUP BY agent_node."agentBlockId"
+        ORDER BY execution_count DESC;
+        """,
+        timestamp_threshold,
     )
 
     # Get the top blocks based on execution count

autogpt_platform/backend/backend/api/features/builder/model.py

@@ -27,6 +27,7 @@ class SearchEntry(BaseModel):
 
 # Suggestions
 class SuggestionsResponse(BaseModel):
+    otto_suggestions: list[str]
     recent_searches: list[SearchEntry]
     providers: list[ProviderName]
     top_blocks: list[BlockInfo]

autogpt_platform/backend/backend/api/features/builder/routes.py

@@ -1,5 +1,5 @@
 import logging
-from typing import Annotated, Sequence, cast, get_args
+from typing import Annotated, Sequence
 
 import fastapi
 from autogpt_libs.auth.dependencies import get_user_id, requires_user
@@ -10,8 +10,6 @@ from backend.util.models import Pagination
 from . import db as builder_db
 from . import model as builder_model
 
-VALID_FILTER_VALUES = get_args(builder_model.FilterType)
-
 logger = logging.getLogger(__name__)
 
 router = fastapi.APIRouter(
@@ -51,6 +49,11 @@ async def get_suggestions(
     Get all suggestions for the Blocks Menu.
     """
     return builder_model.SuggestionsResponse(
+        otto_suggestions=[
+            "What blocks do I need to get started?",
+            "Help me create a list",
+            "Help me feed my data to Google Maps",
+        ],
         recent_searches=await builder_db.get_recent_searches(user_id),
         providers=[
             ProviderName.TWITTER,
@@ -148,7 +151,7 @@ async def get_providers(
 async def search(
     user_id: Annotated[str, fastapi.Security(get_user_id)],
     search_query: Annotated[str | None, fastapi.Query()] = None,
-    filter: Annotated[str | None, fastapi.Query()] = None,
+    filter: Annotated[list[builder_model.FilterType] | None, fastapi.Query()] = None,
     search_id: Annotated[str | None, fastapi.Query()] = None,
     by_creator: Annotated[list[str] | None, fastapi.Query()] = None,
     page: Annotated[int, fastapi.Query()] = 1,
@@ -157,20 +160,9 @@ async def search(
     """
     Search for blocks (including integrations), marketplace agents, and user library agents.
     """
-    # Parse and validate filter parameter
-    filters: list[builder_model.FilterType]
-    if filter:
-        filter_values = [f.strip() for f in filter.split(",")]
-        invalid_filters = [f for f in filter_values if f not in VALID_FILTER_VALUES]
-        if invalid_filters:
-            raise fastapi.HTTPException(
-                status_code=400,
-                detail=f"Invalid filter value(s): {', '.join(invalid_filters)}. "
-                f"Valid values are: {', '.join(VALID_FILTER_VALUES)}",
-            )
-        filters = cast(list[builder_model.FilterType], filter_values)
-    else:
-        filters = [
+    # If no filters are provided, then we will return all types
+    if not filter:
+        filter = [
             "blocks",
             "integrations",
             "marketplace_agents",
@@ -182,7 +174,7 @@ async def search(
     cached_results = await builder_db.get_sorted_search_results(
         user_id=user_id,
         search_query=search_query,
-        filters=filters,
+        filters=filter,
         by_creator=by_creator,
     )
 
@@ -204,7 +196,7 @@ async def search(
         user_id,
         builder_model.SearchEntry(
             search_query=search_query,
-            filter=filters,
+            filter=filter,
             by_creator=by_creator,
             search_id=search_id,
         ),

autogpt_platform/backend/backend/api/features/chat/model.py

@@ -290,6 +290,11 @@ async def _cache_session(session: ChatSession) -> None:
     await async_redis.setex(redis_key, config.session_ttl, session.model_dump_json())
 
 
+async def cache_chat_session(session: ChatSession) -> None:
+    """Cache a chat session without persisting to the database."""
+    await _cache_session(session)
+
+
 async def _get_session_from_db(session_id: str) -> ChatSession | None:
     """Get a chat session from the database."""
     prisma_session = await chat_db.get_chat_session(session_id)

autogpt_platform/backend/backend/api/features/chat/routes.py

@@ -172,12 +172,12 @@ async def get_session(
         user_id: The optional authenticated user ID, or None for anonymous access.
 
     Returns:
-        SessionDetailResponse: Details for the requested session; raises NotFoundError if not found.
+        SessionDetailResponse: Details for the requested session, or None if not found.
 
     """
     session = await get_chat_session(session_id, user_id)
     if not session:
-        raise NotFoundError(f"Session {session_id} not found")
+        raise NotFoundError(f"Session {session_id} not found.")
 
     messages = [message.model_dump() for message in session.messages]
     logger.info(
@@ -222,6 +222,8 @@ async def stream_chat_post(
     session = await _validate_and_get_session(session_id, user_id)
 
     async def event_generator() -> AsyncGenerator[str, None]:
+        chunk_count = 0
+        first_chunk_type: str | None = None
         async for chunk in chat_service.stream_chat_completion(
             session_id,
             request.message,
@@ -230,7 +232,26 @@ async def stream_chat_post(
             session=session,  # Pass pre-fetched session to avoid double-fetch
             context=request.context,
         ):
+            if chunk_count < 3:
+                logger.info(
+                    "Chat stream chunk",
+                    extra={
+                        "session_id": session_id,
+                        "chunk_type": str(chunk.type),
+                    },
+                )
+            if not first_chunk_type:
+                first_chunk_type = str(chunk.type)
+            chunk_count += 1
             yield chunk.to_sse()
+        logger.info(
+            "Chat stream completed",
+            extra={
+                "session_id": session_id,
+                "chunk_count": chunk_count,
+                "first_chunk_type": first_chunk_type,
+            },
+        )
         # AI SDK protocol termination
         yield "data: [DONE]\n\n"
 
@@ -275,6 +296,8 @@ async def stream_chat_get(
     session = await _validate_and_get_session(session_id, user_id)
 
     async def event_generator() -> AsyncGenerator[str, None]:
+        chunk_count = 0
+        first_chunk_type: str | None = None
         async for chunk in chat_service.stream_chat_completion(
             session_id,
             message,
@@ -282,7 +305,26 @@ async def stream_chat_get(
             user_id=user_id,
             session=session,  # Pass pre-fetched session to avoid double-fetch
         ):
+            if chunk_count < 3:
+                logger.info(
+                    "Chat stream chunk",
+                    extra={
+                        "session_id": session_id,
+                        "chunk_type": str(chunk.type),
+                    },
+                )
+            if not first_chunk_type:
+                first_chunk_type = str(chunk.type)
+            chunk_count += 1
             yield chunk.to_sse()
+        logger.info(
+            "Chat stream completed",
+            extra={
+                "session_id": session_id,
+                "chunk_count": chunk_count,
+                "first_chunk_type": first_chunk_type,
+            },
+        )
         # AI SDK protocol termination
         yield "data: [DONE]\n\n"
 

autogpt_platform/backend/backend/api/features/chat/service.py

@@ -1,15 +1,18 @@
 import asyncio
 import logging
+import time
+from asyncio import CancelledError
 from collections.abc import AsyncGenerator
 from typing import Any
 
 import orjson
-from langfuse import Langfuse
+from langfuse import get_client, propagate_attributes
+from langfuse.openai import openai  # type: ignore
 from openai import (
     APIConnectionError,
     APIError,
     APIStatusError,
-    AsyncOpenAI,
+    PermissionDeniedError,
     RateLimitError,
 )
 from openai.types.chat import ChatCompletionChunk, ChatCompletionToolParam
@@ -21,12 +24,12 @@ from backend.data.understanding import (
 from backend.util.exceptions import NotFoundError
 from backend.util.settings import Settings
 
-from . import db as chat_db
 from .config import ChatConfig
 from .model import (
     ChatMessage,
     ChatSession,
     Usage,
+    cache_chat_session,
     get_chat_session,
     update_session_title,
     upsert_chat_session,
@@ -50,10 +53,10 @@ logger = logging.getLogger(__name__)
 
 config = ChatConfig()
 settings = Settings()
-client = AsyncOpenAI(api_key=config.api_key, base_url=config.base_url)
+client = openai.AsyncOpenAI(api_key=config.api_key, base_url=config.base_url)
 
-# Langfuse client (lazy initialization)
-_langfuse_client: Langfuse | None = None
+
+langfuse = get_client()
 
 
 class LangfuseNotConfiguredError(Exception):
@@ -69,65 +72,6 @@ def _is_langfuse_configured() -> bool:
     )
 
 
-def _get_langfuse_client() -> Langfuse:
-    """Get or create the Langfuse client for prompt management and tracing."""
-    global _langfuse_client
-    if _langfuse_client is None:
-        if not _is_langfuse_configured():
-            raise LangfuseNotConfiguredError(
-                "Langfuse is not configured. The chat feature requires Langfuse for prompt management. "
-                "Please set the LANGFUSE_PUBLIC_KEY and LANGFUSE_SECRET_KEY environment variables."
-            )
-        _langfuse_client = Langfuse(
-            public_key=settings.secrets.langfuse_public_key,
-            secret_key=settings.secrets.langfuse_secret_key,
-            host=settings.secrets.langfuse_host or "https://cloud.langfuse.com",
-        )
-    return _langfuse_client
-
-
-def _get_environment() -> str:
-    """Get the current environment name for Langfuse tagging."""
-    return settings.config.app_env.value
-
-
-def _get_langfuse_prompt() -> str:
-    """Fetch the latest production prompt from Langfuse.
-
-    Returns:
-        The compiled prompt text from Langfuse.
-
-    Raises:
-        Exception: If Langfuse is unavailable or prompt fetch fails.
-    """
-    try:
-        langfuse = _get_langfuse_client()
-        # cache_ttl_seconds=0 disables SDK caching to always get the latest prompt
-        prompt = langfuse.get_prompt(config.langfuse_prompt_name, cache_ttl_seconds=0)
-        compiled = prompt.compile()
-        logger.info(
-            f"Fetched prompt '{config.langfuse_prompt_name}' from Langfuse "
-            f"(version: {prompt.version})"
-        )
-        return compiled
-    except Exception as e:
-        logger.error(f"Failed to fetch prompt from Langfuse: {e}")
-        raise
-
-
-async def _is_first_session(user_id: str) -> bool:
-    """Check if this is the user's first chat session.
-
-    Returns True if the user has 1 or fewer sessions (meaning this is their first).
-    """
-    try:
-        session_count = await chat_db.get_user_session_count(user_id)
-        return session_count <= 1
-    except Exception as e:
-        logger.warning(f"Failed to check session count for user {user_id}: {e}")
-        return False  # Default to non-onboarding if we can't check
-
-
 async def _build_system_prompt(user_id: str | None) -> tuple[str, Any]:
     """Build the full system prompt including business understanding if available.
 
@@ -139,8 +83,6 @@ async def _build_system_prompt(user_id: str | None) -> tuple[str, Any]:
         Tuple of (compiled prompt string, Langfuse prompt object for tracing)
     """
 
-    langfuse = _get_langfuse_client()
-
     # cache_ttl_seconds=0 disables SDK caching to always get the latest prompt
     prompt = langfuse.get_prompt(config.langfuse_prompt_name, cache_ttl_seconds=0)
 
@@ -158,7 +100,7 @@ async def _build_system_prompt(user_id: str | None) -> tuple[str, Any]:
         context = "This is the first time you are meeting the user. Greet them and introduce them to the platform"
 
     compiled = prompt.compile(users_information=context)
-    return compiled, prompt
+    return compiled, understanding
 
 
 async def _generate_session_title(message: str) -> str | None:
@@ -217,6 +159,7 @@ async def assign_user_to_session(
 async def stream_chat_completion(
     session_id: str,
     message: str | None = None,
+    tool_call_response: str | None = None,
     is_user_message: bool = True,
     user_id: str | None = None,
     retry_count: int = 0,
@@ -256,11 +199,6 @@ async def stream_chat_completion(
         yield StreamFinish()
         return
 
-    # Langfuse observations will be created after session is loaded (need messages for input)
-    # Initialize to None so finally block can safely check and end them
-    trace = None
-    generation = None
-
     # Only fetch from Redis if session not provided (initial call)
     if session is None:
         session = await get_chat_session(session_id, user_id)
@@ -336,297 +274,349 @@ async def stream_chat_completion(
             asyncio.create_task(_update_title())
 
     # Build system prompt with business understanding
-    system_prompt, langfuse_prompt = await _build_system_prompt(user_id)
-
-    # Build input messages including system prompt for complete Langfuse logging
-    trace_input_messages = [{"role": "system", "content": system_prompt}] + [
-        m.model_dump() for m in session.messages
-    ]
+    system_prompt, understanding = await _build_system_prompt(user_id)
 
     # Create Langfuse trace for this LLM call (each call gets its own trace, grouped by session_id)
     # Using v3 SDK: start_observation creates a root span, update_trace sets trace-level attributes
-    try:
-        langfuse = _get_langfuse_client()
-        env = _get_environment()
-        trace = langfuse.start_observation(
-            name="chat_completion",
-            input={"messages": trace_input_messages},
-            metadata={
-                "environment": env,
-                "model": config.model,
-                "message_count": len(session.messages),
-                "prompt_name": langfuse_prompt.name if langfuse_prompt else None,
-                "prompt_version": langfuse_prompt.version if langfuse_prompt else None,
-            },
-        )
-        # Set trace-level attributes (session_id, user_id, tags)
-        trace.update_trace(
+    input = message
+    if not message and tool_call_response:
+        input = tool_call_response
+
+    langfuse = get_client()
+    with langfuse.start_as_current_observation(
+        as_type="span",
+        name="user-copilot-request",
+        input=input,
+    ) as span:
+        with propagate_attributes(
             session_id=session_id,
             user_id=user_id,
-            tags=[env, "copilot"],
-        )
-    except Exception as e:
-        logger.warning(f"Failed to create Langfuse trace: {e}")
+            tags=["copilot"],
+            metadata={
+                "users_information": format_understanding_for_prompt(understanding)[
+                    :200
+                ]  # langfuse only accepts upto to 200 chars
+            },
+        ):
 
-    # Initialize variables that will be used in finally block (must be defined before try)
-    assistant_response = ChatMessage(
-        role="assistant",
-        content="",
-    )
-    accumulated_tool_calls: list[dict[str, Any]] = []
-
-    # Wrap main logic in try/finally to ensure Langfuse observations are always ended
-    try:
-        has_yielded_end = False
-        has_yielded_error = False
-        has_done_tool_call = False
-        has_received_text = False
-        text_streaming_ended = False
-        tool_response_messages: list[ChatMessage] = []
-        should_retry = False
-
-        # Generate unique IDs for AI SDK protocol
-        import uuid as uuid_module
-
-        message_id = str(uuid_module.uuid4())
-        text_block_id = str(uuid_module.uuid4())
-
-        # Yield message start
-        yield StreamStart(messageId=message_id)
-
-        # Create Langfuse generation for each LLM call, linked to the prompt
-        # Using v3 SDK: start_observation with as_type="generation"
-        generation = (
-            trace.start_observation(
-                as_type="generation",
-                name="llm_call",
-                model=config.model,
-                input={"messages": trace_input_messages},
-                prompt=langfuse_prompt,
+            # Initialize variables that will be used in finally block (must be defined before try)
+            assistant_response = ChatMessage(
+                role="assistant",
+                content="",
             )
-            if trace
-            else None
-        )
+            accumulated_tool_calls: list[dict[str, Any]] = []
+            has_saved_assistant_message = False
+            has_appended_streaming_message = False
+            last_cache_time = 0.0
+            last_cache_content_len = 0
 
-        try:
-            async for chunk in _stream_chat_chunks(
-                session=session,
-                tools=tools,
-                system_prompt=system_prompt,
-                text_block_id=text_block_id,
-            ):
+            # Wrap main logic in try/finally to ensure Langfuse observations are always ended
+            has_yielded_end = False
+            has_yielded_error = False
+            has_done_tool_call = False
+            has_received_text = False
+            text_streaming_ended = False
+            tool_response_messages: list[ChatMessage] = []
+            should_retry = False
 
-                if isinstance(chunk, StreamTextStart):
-                    # Emit text-start before first text delta
-                    if not has_received_text:
+            # Generate unique IDs for AI SDK protocol
+            import uuid as uuid_module
+
+            message_id = str(uuid_module.uuid4())
+            text_block_id = str(uuid_module.uuid4())
+
+            # Yield message start
+            yield StreamStart(messageId=message_id)
+
+            try:
+                async for chunk in _stream_chat_chunks(
+                    session=session,
+                    tools=tools,
+                    system_prompt=system_prompt,
+                    text_block_id=text_block_id,
+                ):
+
+                    if isinstance(chunk, StreamTextStart):
+                        # Emit text-start before first text delta
+                        if not has_received_text:
+                            yield chunk
+                    elif isinstance(chunk, StreamTextDelta):
+                        delta = chunk.delta or ""
+                        assert assistant_response.content is not None
+                        assistant_response.content += delta
+                        has_received_text = True
+                        if not has_appended_streaming_message:
+                            session.messages.append(assistant_response)
+                            has_appended_streaming_message = True
+                        current_time = time.monotonic()
+                        content_len = len(assistant_response.content)
+                        if (
+                            current_time - last_cache_time >= 1.0
+                            and content_len > last_cache_content_len
+                        ):
+                            try:
+                                await cache_chat_session(session)
+                            except Exception as e:
+                                logger.warning(
+                                    f"Failed to cache partial session {session.session_id}: {e}"
+                                )
+                            last_cache_time = current_time
+                            last_cache_content_len = content_len
                         yield chunk
-                elif isinstance(chunk, StreamTextDelta):
-                    delta = chunk.delta or ""
-                    assert assistant_response.content is not None
-                    assistant_response.content += delta
-                    has_received_text = True
-                    yield chunk
-                elif isinstance(chunk, StreamTextEnd):
-                    # Emit text-end after text completes
-                    if has_received_text and not text_streaming_ended:
-                        text_streaming_ended = True
-                        yield chunk
-                elif isinstance(chunk, StreamToolInputStart):
-                    # Emit text-end before first tool call, but only if we've received text
-                    if has_received_text and not text_streaming_ended:
-                        yield StreamTextEnd(id=text_block_id)
-                        text_streaming_ended = True
-                    yield chunk
-                elif isinstance(chunk, StreamToolInputAvailable):
-                    # Accumulate tool calls in OpenAI format
-                    accumulated_tool_calls.append(
-                        {
-                            "id": chunk.toolCallId,
-                            "type": "function",
-                            "function": {
-                                "name": chunk.toolName,
-                                "arguments": orjson.dumps(chunk.input).decode("utf-8"),
-                            },
-                        }
-                    )
-                elif isinstance(chunk, StreamToolOutputAvailable):
-                    result_content = (
-                        chunk.output
-                        if isinstance(chunk.output, str)
-                        else orjson.dumps(chunk.output).decode("utf-8")
-                    )
-                    tool_response_messages.append(
-                        ChatMessage(
-                            role="tool",
-                            content=result_content,
-                            tool_call_id=chunk.toolCallId,
-                        )
-                    )
-                    has_done_tool_call = True
-                    # Track if any tool execution failed
-                    if not chunk.success:
-                        logger.warning(
-                            f"Tool {chunk.toolName} (ID: {chunk.toolCallId}) execution failed"
-                        )
-                    yield chunk
-                elif isinstance(chunk, StreamFinish):
-                    if not has_done_tool_call:
-                        # Emit text-end before finish if we received text but haven't closed it
+                    elif isinstance(chunk, StreamTextEnd):
+                        # Emit text-end after text completes
+                        if has_received_text and not text_streaming_ended:
+                            text_streaming_ended = True
+                            if assistant_response.content:
+                                logger.warn(
+                                    f"StreamTextEnd: Attempting to set output {assistant_response.content}"
+                                )
+                                span.update_trace(output=assistant_response.content)
+                                span.update(output=assistant_response.content)
+                            yield chunk
+                    elif isinstance(chunk, StreamToolInputStart):
+                        # Emit text-end before first tool call, but only if we've received text
                         if has_received_text and not text_streaming_ended:
                             yield StreamTextEnd(id=text_block_id)
                             text_streaming_ended = True
-                        has_yielded_end = True
                         yield chunk
-                elif isinstance(chunk, StreamError):
-                    has_yielded_error = True
-                elif isinstance(chunk, StreamUsage):
-                    session.usage.append(
-                        Usage(
-                            prompt_tokens=chunk.promptTokens,
-                            completion_tokens=chunk.completionTokens,
-                            total_tokens=chunk.totalTokens,
+                    elif isinstance(chunk, StreamToolInputAvailable):
+                        # Accumulate tool calls in OpenAI format
+                        accumulated_tool_calls.append(
+                            {
+                                "id": chunk.toolCallId,
+                                "type": "function",
+                                "function": {
+                                    "name": chunk.toolName,
+                                    "arguments": orjson.dumps(chunk.input).decode(
+                                        "utf-8"
+                                    ),
+                                },
+                            }
                         )
-                    )
-                else:
-                    logger.error(f"Unknown chunk type: {type(chunk)}", exc_info=True)
-        except Exception as e:
-            logger.error(f"Error during stream: {e!s}", exc_info=True)
+                    elif isinstance(chunk, StreamToolOutputAvailable):
+                        result_content = (
+                            chunk.output
+                            if isinstance(chunk.output, str)
+                            else orjson.dumps(chunk.output).decode("utf-8")
+                        )
+                        tool_response_messages.append(
+                            ChatMessage(
+                                role="tool",
+                                content=result_content,
+                                tool_call_id=chunk.toolCallId,
+                            )
+                        )
+                        has_done_tool_call = True
+                        # Track if any tool execution failed
+                        if not chunk.success:
+                            logger.warning(
+                                f"Tool {chunk.toolName} (ID: {chunk.toolCallId}) execution failed"
+                            )
+                        yield chunk
+                    elif isinstance(chunk, StreamFinish):
+                        if not has_done_tool_call:
+                            # Emit text-end before finish if we received text but haven't closed it
+                            if has_received_text and not text_streaming_ended:
+                                yield StreamTextEnd(id=text_block_id)
+                                text_streaming_ended = True
 
-            # Check if this is a retryable error (JSON parsing, incomplete tool calls, etc.)
-            is_retryable = isinstance(e, (orjson.JSONDecodeError, KeyError, TypeError))
+                            # Save assistant message before yielding finish to ensure it's persisted
+                            # even if client disconnects immediately after receiving StreamFinish
+                            if not has_saved_assistant_message:
+                                messages_to_save_early: list[ChatMessage] = []
+                                if accumulated_tool_calls:
+                                    assistant_response.tool_calls = (
+                                        accumulated_tool_calls
+                                    )
+                                if not has_appended_streaming_message and (
+                                    assistant_response.content
+                                    or assistant_response.tool_calls
+                                ):
+                                    messages_to_save_early.append(assistant_response)
+                                messages_to_save_early.extend(tool_response_messages)
 
-            if is_retryable and retry_count < config.max_retries:
-                logger.info(
-                    f"Retryable error encountered. Attempt {retry_count + 1}/{config.max_retries}"
+                                if messages_to_save_early:
+                                    session.messages.extend(messages_to_save_early)
+                                    logger.info(
+                                        f"Saving assistant message before StreamFinish: "
+                                        f"content_len={len(assistant_response.content or '')}, "
+                                        f"tool_calls={len(assistant_response.tool_calls or [])}, "
+                                        f"tool_responses={len(tool_response_messages)}"
+                                    )
+                                if (
+                                    messages_to_save_early
+                                    or has_appended_streaming_message
+                                ):
+                                    await upsert_chat_session(session)
+                                    has_saved_assistant_message = True
+
+                            has_yielded_end = True
+                            yield chunk
+                    elif isinstance(chunk, StreamError):
+                        has_yielded_error = True
+                        yield chunk
+                    elif isinstance(chunk, StreamUsage):
+                        session.usage.append(
+                            Usage(
+                                prompt_tokens=chunk.promptTokens,
+                                completion_tokens=chunk.completionTokens,
+                                total_tokens=chunk.totalTokens,
+                            )
+                        )
+                    else:
+                        logger.error(
+                            f"Unknown chunk type: {type(chunk)}", exc_info=True
+                        )
+                if assistant_response.content:
+                    langfuse.update_current_trace(output=assistant_response.content)
+                    langfuse.update_current_span(output=assistant_response.content)
+                elif tool_response_messages:
+                    langfuse.update_current_trace(output=str(tool_response_messages))
+                    langfuse.update_current_span(output=str(tool_response_messages))
+
+            except CancelledError:
+                if not has_saved_assistant_message:
+                    if accumulated_tool_calls:
+                        assistant_response.tool_calls = accumulated_tool_calls
+                    if assistant_response.content:
+                        assistant_response.content = (
+                            f"{assistant_response.content}\n\n[interrupted]"
+                        )
+                    else:
+                        assistant_response.content = "[interrupted]"
+                    if not has_appended_streaming_message:
+                        session.messages.append(assistant_response)
+                    if tool_response_messages:
+                        session.messages.extend(tool_response_messages)
+                    try:
+                        await upsert_chat_session(session)
+                    except Exception as e:
+                        logger.warning(
+                            f"Failed to save interrupted session {session.session_id}: {e}"
+                        )
+                raise
+            except Exception as e:
+                logger.error(f"Error during stream: {e!s}", exc_info=True)
+
+                # Check if this is a retryable error (JSON parsing, incomplete tool calls, etc.)
+                is_retryable = isinstance(
+                    e, (orjson.JSONDecodeError, KeyError, TypeError)
                 )
-                should_retry = True
-            else:
-                # Non-retryable error or max retries exceeded
-                # Save any partial progress before reporting error
+
+                if is_retryable and retry_count < config.max_retries:
+                    logger.info(
+                        f"Retryable error encountered. Attempt {retry_count + 1}/{config.max_retries}"
+                    )
+                    should_retry = True
+                else:
+                    # Non-retryable error or max retries exceeded
+                    # Save any partial progress before reporting error
+                    messages_to_save: list[ChatMessage] = []
+
+                    # Add assistant message if it has content or tool calls
+                    if accumulated_tool_calls:
+                        assistant_response.tool_calls = accumulated_tool_calls
+                    if not has_appended_streaming_message and (
+                        assistant_response.content or assistant_response.tool_calls
+                    ):
+                        messages_to_save.append(assistant_response)
+
+                    # Add tool response messages after assistant message
+                    messages_to_save.extend(tool_response_messages)
+
+                    if not has_saved_assistant_message:
+                        if messages_to_save:
+                            session.messages.extend(messages_to_save)
+                        if messages_to_save or has_appended_streaming_message:
+                            await upsert_chat_session(session)
+
+                    if not has_yielded_error:
+                        error_message = str(e)
+                        if not is_retryable:
+                            error_message = f"Non-retryable error: {error_message}"
+                        elif retry_count >= config.max_retries:
+                            error_message = f"Max retries ({config.max_retries}) exceeded: {error_message}"
+
+                        error_response = StreamError(errorText=error_message)
+                        yield error_response
+                    if not has_yielded_end:
+                        yield StreamFinish()
+                    return
+
+            # Handle retry outside of exception handler to avoid nesting
+            if should_retry and retry_count < config.max_retries:
+                logger.info(
+                    f"Retrying stream_chat_completion for session {session_id}, attempt {retry_count + 1}"
+                )
+                async for chunk in stream_chat_completion(
+                    session_id=session.session_id,
+                    user_id=user_id,
+                    retry_count=retry_count + 1,
+                    session=session,
+                    context=context,
+                ):
+                    yield chunk
+                return  # Exit after retry to avoid double-saving in finally block
+
+            # Normal completion path - save session and handle tool call continuation
+            # Only save if we haven't already saved when StreamFinish was received
+            if not has_saved_assistant_message:
+                logger.info(
+                    f"Normal completion path: session={session.session_id}, "
+                    f"current message_count={len(session.messages)}"
+                )
+
+                # Build the messages list in the correct order
                 messages_to_save: list[ChatMessage] = []
 
-                # Add assistant message if it has content or tool calls
+                # Add assistant message with tool_calls if any
                 if accumulated_tool_calls:
                     assistant_response.tool_calls = accumulated_tool_calls
-                if assistant_response.content or assistant_response.tool_calls:
+                    logger.info(
+                        f"Added {len(accumulated_tool_calls)} tool calls to assistant message"
+                    )
+                if not has_appended_streaming_message and (
+                    assistant_response.content or assistant_response.tool_calls
+                ):
                     messages_to_save.append(assistant_response)
+                    logger.info(
+                        f"Saving assistant message with content_len={len(assistant_response.content or '')}, tool_calls={len(assistant_response.tool_calls or [])}"
+                    )
 
                 # Add tool response messages after assistant message
                 messages_to_save.extend(tool_response_messages)
-
-                session.messages.extend(messages_to_save)
-                await upsert_chat_session(session)
-
-                if not has_yielded_error:
-                    error_message = str(e)
-                    if not is_retryable:
-                        error_message = f"Non-retryable error: {error_message}"
-                    elif retry_count >= config.max_retries:
-                        error_message = f"Max retries ({config.max_retries}) exceeded: {error_message}"
-
-                    error_response = StreamError(errorText=error_message)
-                    yield error_response
-                if not has_yielded_end:
-                    yield StreamFinish()
-                return
-
-        # Handle retry outside of exception handler to avoid nesting
-        if should_retry and retry_count < config.max_retries:
-            logger.info(
-                f"Retrying stream_chat_completion for session {session_id}, attempt {retry_count + 1}"
-            )
-            async for chunk in stream_chat_completion(
-                session_id=session.session_id,
-                user_id=user_id,
-                retry_count=retry_count + 1,
-                session=session,
-                context=context,
-            ):
-                yield chunk
-            return  # Exit after retry to avoid double-saving in finally block
-
-        # Normal completion path - save session and handle tool call continuation
-        logger.info(
-            f"Normal completion path: session={session.session_id}, "
-            f"current message_count={len(session.messages)}"
-        )
-
-        # Build the messages list in the correct order
-        messages_to_save: list[ChatMessage] = []
-
-        # Add assistant message with tool_calls if any
-        if accumulated_tool_calls:
-            assistant_response.tool_calls = accumulated_tool_calls
-            logger.info(
-                f"Added {len(accumulated_tool_calls)} tool calls to assistant message"
-            )
-        if assistant_response.content or assistant_response.tool_calls:
-            messages_to_save.append(assistant_response)
-            logger.info(
-                f"Saving assistant message with content_len={len(assistant_response.content or '')}, tool_calls={len(assistant_response.tool_calls or [])}"
-            )
-
-        # Add tool response messages after assistant message
-        messages_to_save.extend(tool_response_messages)
-        logger.info(
-            f"Saving {len(tool_response_messages)} tool response messages, "
-            f"total_to_save={len(messages_to_save)}"
-        )
-
-        session.messages.extend(messages_to_save)
-        logger.info(
-            f"Extended session messages, new message_count={len(session.messages)}"
-        )
-        await upsert_chat_session(session)
-
-        # If we did a tool call, stream the chat completion again to get the next response
-        if has_done_tool_call:
-            logger.info(
-                "Tool call executed, streaming chat completion again to get assistant response"
-            )
-            async for chunk in stream_chat_completion(
-                session_id=session.session_id,
-                user_id=user_id,
-                session=session,  # Pass session object to avoid Redis refetch
-                context=context,
-            ):
-                yield chunk
-
-    finally:
-        # Always end Langfuse observations to prevent resource leaks
-        # Guard against None and catch errors to avoid masking original exceptions
-        if generation is not None:
-            try:
-                latest_usage = session.usage[-1] if session.usage else None
-                generation.update(
-                    model=config.model,
-                    output={
-                        "content": assistant_response.content,
-                        "tool_calls": accumulated_tool_calls or None,
-                    },
-                    usage_details=(
-                        {
-                            "input": latest_usage.prompt_tokens,
-                            "output": latest_usage.completion_tokens,
-                            "total": latest_usage.total_tokens,
-                        }
-                        if latest_usage
-                        else None
-                    ),
+                logger.info(
+                    f"Saving {len(tool_response_messages)} tool response messages, "
+                    f"total_to_save={len(messages_to_save)}"
                 )
-                generation.end()
-            except Exception as e:
-                logger.warning(f"Failed to end Langfuse generation: {e}")
 
-        if trace is not None:
-            try:
-                if accumulated_tool_calls:
-                    trace.update_trace(output={"tool_calls": accumulated_tool_calls})
-                else:
-                    trace.update_trace(output={"response": assistant_response.content})
-                trace.end()
-            except Exception as e:
-                logger.warning(f"Failed to end Langfuse trace: {e}")
+                if messages_to_save:
+                    session.messages.extend(messages_to_save)
+                    logger.info(
+                        f"Extended session messages, new message_count={len(session.messages)}"
+                    )
+                if messages_to_save or has_appended_streaming_message:
+                    await upsert_chat_session(session)
+            else:
+                logger.info(
+                    "Assistant message already saved when StreamFinish was received, "
+                    "skipping duplicate save"
+                )
+
+            # If we did a tool call, stream the chat completion again to get the next response
+            if has_done_tool_call:
+                logger.info(
+                    "Tool call executed, streaming chat completion again to get assistant response"
+                )
+                async for chunk in stream_chat_completion(
+                    session_id=session.session_id,
+                    user_id=user_id,
+                    session=session,  # Pass session object to avoid Redis refetch
+                    context=context,
+                    tool_call_response=str(tool_response_messages),
+                ):
+                    yield chunk
 
 
 # Retry configuration for OpenAI API calls
@@ -654,6 +644,12 @@ def _is_retryable_error(error: Exception) -> bool:
     return False
 
 
+def _is_region_blocked_error(error: Exception) -> bool:
+    if isinstance(error, PermissionDeniedError):
+        return "not available in your region" in str(error).lower()
+    return "not available in your region" in str(error).lower()
+
+
 async def _stream_chat_chunks(
     session: ChatSession,
     tools: list[ChatCompletionToolParam],
@@ -846,7 +842,18 @@ async def _stream_chat_chunks(
                         f"Error in stream (not retrying): {e!s}",
                         exc_info=True,
                     )
-                    error_response = StreamError(errorText=str(e))
+                    error_code = None
+                    error_text = str(e)
+                    if _is_region_blocked_error(e):
+                        error_code = "MODEL_NOT_AVAILABLE_REGION"
+                        error_text = (
+                            "This model is not available in your region. "
+                            "Please connect via VPN and try again."
+                        )
+                    error_response = StreamError(
+                        errorText=error_text,
+                        code=error_code,
+                    )
                     yield error_response
                     yield StreamFinish()
                     return
@@ -900,5 +907,4 @@ async def _yield_tool_call(
         session=session,
     )
 
-    logger.info(f"Yielding Tool execution response: {tool_execution_response}")
     yield tool_execution_response

autogpt_platform/backend/backend/api/features/chat/tools/__init__.py

@@ -30,7 +30,7 @@ TOOL_REGISTRY: dict[str, BaseTool] = {
     "find_library_agent": FindLibraryAgentTool(),
     "run_agent": RunAgentTool(),
     "run_block": RunBlockTool(),
-    "agent_output": AgentOutputTool(),
+    "view_agent_output": AgentOutputTool(),
     "search_docs": SearchDocsTool(),
     "get_doc_page": GetDocPageTool(),
 }

autogpt_platform/backend/backend/api/features/chat/tools/add_understanding.py

@@ -3,6 +3,8 @@
 import logging
 from typing import Any
 
+from langfuse import observe
+
 from backend.api.features.chat.model import ChatSession
 from backend.data.understanding import (
     BusinessUnderstandingInput,
@@ -59,6 +61,7 @@ and automations for the user's specific needs."""
         """Requires authentication to store user-specific data."""
         return True
 
+    @observe(as_type="tool", name="add_understanding")
     async def _execute(
         self,
         user_id: str | None,

autogpt_platform/backend/backend/api/features/chat/tools/agent_generator/core.py

@@ -218,6 +218,7 @@ async def save_agent_to_library(
     library_agents = await library_db.create_library_agent(
         graph=created_graph,
         user_id=user_id,
+        sensitive_action_safe_mode=True,
         create_library_agents_for_sub_graphs=False,
     )
 

autogpt_platform/backend/backend/api/features/chat/tools/agent_output.py

@@ -5,6 +5,7 @@ import re
 from datetime import datetime, timedelta, timezone
 from typing import Any
 
+from langfuse import observe
 from pydantic import BaseModel, field_validator
 
 from backend.api.features.chat.model import ChatSession
@@ -103,7 +104,7 @@ class AgentOutputTool(BaseTool):
 
     @property
     def name(self) -> str:
-        return "agent_output"
+        return "view_agent_output"
 
     @property
     def description(self) -> str:
@@ -328,6 +329,7 @@ class AgentOutputTool(BaseTool):
             total_executions=len(available_executions) if available_executions else 1,
         )
 
+    @observe(as_type="tool", name="view_agent_output")
     async def _execute(
         self,
         user_id: str | None,

autogpt_platform/backend/backend/api/features/chat/tools/create_agent.py

@@ -3,6 +3,8 @@
 import logging
 from typing import Any
 
+from langfuse import observe
+
 from backend.api.features.chat.model import ChatSession
 
 from .agent_generator import (
@@ -78,6 +80,7 @@ class CreateAgentTool(BaseTool):
             "required": ["description"],
         }
 
+    @observe(as_type="tool", name="create_agent")
     async def _execute(
         self,
         user_id: str | None,

autogpt_platform/backend/backend/api/features/chat/tools/edit_agent.py

@@ -3,6 +3,8 @@
 import logging
 from typing import Any
 
+from langfuse import observe
+
 from backend.api.features.chat.model import ChatSession
 
 from .agent_generator import (
@@ -85,6 +87,7 @@ class EditAgentTool(BaseTool):
             "required": ["agent_id", "changes"],
         }
 
+    @observe(as_type="tool", name="edit_agent")
     async def _execute(
         self,
         user_id: str | None,

autogpt_platform/backend/backend/api/features/chat/tools/find_agent.py

@@ -2,6 +2,8 @@
 
 from typing import Any
 
+from langfuse import observe
+
 from backend.api.features.chat.model import ChatSession
 
 from .agent_search import search_agents
@@ -35,6 +37,7 @@ class FindAgentTool(BaseTool):
             "required": ["query"],
         }
 
+    @observe(as_type="tool", name="find_agent")
     async def _execute(
         self, user_id: str | None, session: ChatSession, **kwargs
     ) -> ToolResponseBase:

autogpt_platform/backend/backend/api/features/chat/tools/find_block.py

@@ -1,6 +1,7 @@
 import logging
 from typing import Any
 
+from langfuse import observe
 from prisma.enums import ContentType
 
 from backend.api.features.chat.model import ChatSession
@@ -55,6 +56,7 @@ class FindBlockTool(BaseTool):
     def requires_auth(self) -> bool:
         return True
 
+    @observe(as_type="tool", name="find_block")
     async def _execute(
         self,
         user_id: str | None,

autogpt_platform/backend/backend/api/features/chat/tools/find_library_agent.py

@@ -2,6 +2,8 @@
 
 from typing import Any
 
+from langfuse import observe
+
 from backend.api.features.chat.model import ChatSession
 
 from .agent_search import search_agents
@@ -41,6 +43,7 @@ class FindLibraryAgentTool(BaseTool):
     def requires_auth(self) -> bool:
         return True
 
+    @observe(as_type="tool", name="find_library_agent")
     async def _execute(
         self, user_id: str | None, session: ChatSession, **kwargs
     ) -> ToolResponseBase:

autogpt_platform/backend/backend/api/features/chat/tools/get_doc_page.py

@@ -4,6 +4,8 @@ import logging
 from pathlib import Path
 from typing import Any
 
+from langfuse import observe
+
 from backend.api.features.chat.model import ChatSession
 from backend.api.features.chat.tools.base import BaseTool
 from backend.api.features.chat.tools.models import (
@@ -71,6 +73,7 @@ class GetDocPageTool(BaseTool):
         url_path = path.rsplit(".", 1)[0] if "." in path else path
         return f"{DOCS_BASE_URL}/{url_path}"
 
+    @observe(as_type="tool", name="get_doc_page")
     async def _execute(
         self,
         user_id: str | None,

autogpt_platform/backend/backend/api/features/chat/tools/run_agent.py

@@ -3,6 +3,7 @@
 import logging
 from typing import Any
 
+from langfuse import observe
 from pydantic import BaseModel, Field, field_validator
 
 from backend.api.features.chat.config import ChatConfig
@@ -32,7 +33,7 @@ from .models import (
     UserReadiness,
 )
 from .utils import (
-    check_user_has_required_credentials,
+    build_missing_credentials_from_graph,
     extract_credentials_from_schema,
     fetch_graph_from_store_slug,
     get_or_create_library_agent,
@@ -154,6 +155,7 @@ class RunAgentTool(BaseTool):
         """All operations require authentication."""
         return True
 
+    @observe(as_type="tool", name="run_agent")
     async def _execute(
         self,
         user_id: str | None,
@@ -235,15 +237,13 @@ class RunAgentTool(BaseTool):
                 # Return credentials needed response with input data info
                 # The UI handles credential setup automatically, so the message
                 # focuses on asking about input data
-                credentials = extract_credentials_from_schema(
-                    graph.credentials_input_schema
+                requirements_creds_dict = build_missing_credentials_from_graph(
+                    graph, None
                 )
-                missing_creds_check = await check_user_has_required_credentials(
-                    user_id, credentials
+                missing_credentials_dict = build_missing_credentials_from_graph(
+                    graph, graph_credentials
                 )
-                missing_credentials_dict = {
-                    c.id: c.model_dump() for c in missing_creds_check
-                }
+                requirements_creds_list = list(requirements_creds_dict.values())
 
                 return SetupRequirementsResponse(
                     message=self._build_inputs_message(graph, MSG_WHAT_VALUES_TO_USE),
@@ -257,7 +257,7 @@ class RunAgentTool(BaseTool):
                             ready_to_run=False,
                         ),
                         requirements={
-                            "credentials": [c.model_dump() for c in credentials],
+                            "credentials": requirements_creds_list,
                             "inputs": self._get_inputs_list(graph.input_schema),
                             "execution_modes": self._get_execution_modes(graph),
                         },

autogpt_platform/backend/backend/api/features/chat/tools/run_block.py

@@ -4,6 +4,8 @@ import logging
 from collections import defaultdict
 from typing import Any
 
+from langfuse import observe
+
 from backend.api.features.chat.model import ChatSession
 from backend.data.block import get_block
 from backend.data.execution import ExecutionContext
@@ -20,6 +22,7 @@ from .models import (
     ToolResponseBase,
     UserReadiness,
 )
+from .utils import build_missing_credentials_from_field_info
 
 logger = logging.getLogger(__name__)
 
@@ -127,6 +130,7 @@ class RunBlockTool(BaseTool):
 
         return matched_credentials, missing_credentials
 
+    @observe(as_type="tool", name="run_block")
     async def _execute(
         self,
         user_id: str | None,
@@ -186,7 +190,11 @@ class RunBlockTool(BaseTool):
 
         if missing_credentials:
             # Return setup requirements response with missing credentials
-            missing_creds_dict = {c.id: c.model_dump() for c in missing_credentials}
+            credentials_fields_info = block.input_schema.get_credentials_fields_info()
+            missing_creds_dict = build_missing_credentials_from_field_info(
+                credentials_fields_info, set(matched_credentials.keys())
+            )
+            missing_creds_list = list(missing_creds_dict.values())
 
             return SetupRequirementsResponse(
                 message=(
@@ -203,7 +211,7 @@ class RunBlockTool(BaseTool):
                         ready_to_run=False,
                     ),
                     requirements={
-                        "credentials": [c.model_dump() for c in missing_credentials],
+                        "credentials": missing_creds_list,
                         "inputs": self._get_inputs_list(block),
                         "execution_modes": ["immediate"],
                     },

autogpt_platform/backend/backend/api/features/chat/tools/search_docs.py

@@ -3,6 +3,7 @@
 import logging
 from typing import Any
 
+from langfuse import observe
 from prisma.enums import ContentType
 
 from backend.api.features.chat.model import ChatSession
@@ -87,6 +88,7 @@ class SearchDocsTool(BaseTool):
         url_path = path.rsplit(".", 1)[0] if "." in path else path
         return f"{DOCS_BASE_URL}/{url_path}"
 
+    @observe(as_type="tool", name="search_docs")
     async def _execute(
         self,
         user_id: str | None,

autogpt_platform/backend/backend/api/features/chat/tools/utils.py

@@ -8,7 +8,7 @@ from backend.api.features.library import model as library_model
 from backend.api.features.store import db as store_db
 from backend.data import graph as graph_db
 from backend.data.graph import GraphModel
-from backend.data.model import CredentialsMetaInput
+from backend.data.model import CredentialsFieldInfo, CredentialsMetaInput
 from backend.integrations.creds_manager import IntegrationCredentialsManager
 from backend.util.exceptions import NotFoundError
 
@@ -89,6 +89,59 @@ def extract_credentials_from_schema(
     return credentials
 
 
+def _serialize_missing_credential(
+    field_key: str, field_info: CredentialsFieldInfo
+) -> dict[str, Any]:
+    """
+    Convert credential field info into a serializable dict that preserves all supported
+    credential types (e.g., api_key + oauth2) so the UI can offer multiple options.
+    """
+    supported_types = sorted(field_info.supported_types)
+    provider = next(iter(field_info.provider), "unknown")
+    scopes = sorted(field_info.required_scopes or [])
+
+    return {
+        "id": field_key,
+        "title": field_key.replace("_", " ").title(),
+        "provider": provider,
+        "provider_name": provider.replace("_", " ").title(),
+        "type": supported_types[0] if supported_types else "api_key",
+        "types": supported_types,
+        "scopes": scopes,
+    }
+
+
+def build_missing_credentials_from_graph(
+    graph: GraphModel, matched_credentials: dict[str, CredentialsMetaInput] | None
+) -> dict[str, Any]:
+    """
+    Build a missing_credentials mapping from a graph's aggregated credentials inputs,
+    preserving all supported credential types for each field.
+    """
+    matched_keys = set(matched_credentials.keys()) if matched_credentials else set()
+    aggregated_fields = graph.aggregate_credentials_inputs()
+
+    return {
+        field_key: _serialize_missing_credential(field_key, field_info)
+        for field_key, (field_info, _node_fields) in aggregated_fields.items()
+        if field_key not in matched_keys
+    }
+
+
+def build_missing_credentials_from_field_info(
+    credential_fields: dict[str, CredentialsFieldInfo],
+    matched_keys: set[str],
+) -> dict[str, Any]:
+    """
+    Build missing_credentials mapping from a simple credentials field info dictionary.
+    """
+    return {
+        field_key: _serialize_missing_credential(field_key, field_info)
+        for field_key, field_info in credential_fields.items()
+        if field_key not in matched_keys
+    }
+
+
 def extract_credentials_as_dict(
     credentials_input_schema: dict[str, Any] | None,
 ) -> dict[str, CredentialsMetaInput]:

autogpt_platform/backend/backend/api/features/library/db.py

@@ -401,27 +401,11 @@ async def add_generated_agent_image(
     )
 
 
-def _initialize_graph_settings(graph: graph_db.GraphModel) -> GraphSettings:
-    """
-    Initialize GraphSettings based on graph content.
-
-    Args:
-        graph: The graph to analyze
-
-    Returns:
-        GraphSettings with appropriate human_in_the_loop_safe_mode value
-    """
-    if graph.has_human_in_the_loop:
-        # Graph has HITL blocks - set safe mode to True by default
-        return GraphSettings(human_in_the_loop_safe_mode=True)
-    else:
-        # Graph has no HITL blocks - keep None
-        return GraphSettings(human_in_the_loop_safe_mode=None)
-
-
 async def create_library_agent(
     graph: graph_db.GraphModel,
     user_id: str,
+    hitl_safe_mode: bool = True,
+    sensitive_action_safe_mode: bool = False,
     create_library_agents_for_sub_graphs: bool = True,
 ) -> list[library_model.LibraryAgent]:
     """
@@ -430,6 +414,8 @@ async def create_library_agent(
     Args:
         agent: The agent/Graph to add to the library.
         user_id: The user to whom the agent will be added.
+        hitl_safe_mode: Whether HITL blocks require manual review (default True).
+        sensitive_action_safe_mode: Whether sensitive action blocks require review.
         create_library_agents_for_sub_graphs: If True, creates LibraryAgent records for sub-graphs as well.
 
     Returns:
@@ -465,7 +451,11 @@ async def create_library_agent(
                             }
                         },
                         settings=SafeJson(
-                            _initialize_graph_settings(graph_entry).model_dump()
+                            GraphSettings.from_graph(
+                                graph_entry,
+                                hitl_safe_mode=hitl_safe_mode,
+                                sensitive_action_safe_mode=sensitive_action_safe_mode,
+                            ).model_dump()
                         ),
                     ),
                     include=library_agent_include(
@@ -627,33 +617,6 @@ async def update_library_agent(
         raise DatabaseError("Failed to update library agent") from e
 
 
-async def update_library_agent_settings(
-    user_id: str,
-    agent_id: str,
-    settings: GraphSettings,
-) -> library_model.LibraryAgent:
-    """
-    Updates the settings for a specific LibraryAgent.
-
-    Args:
-        user_id: The owner of the LibraryAgent.
-        agent_id: The ID of the LibraryAgent to update.
-        settings: New GraphSettings to apply.
-
-    Returns:
-        The updated LibraryAgent.
-
-    Raises:
-        NotFoundError: If the specified LibraryAgent does not exist.
-        DatabaseError: If there's an error in the update operation.
-    """
-    return await update_library_agent(
-        library_agent_id=agent_id,
-        user_id=user_id,
-        settings=settings,
-    )
-
-
 async def delete_library_agent(
     library_agent_id: str, user_id: str, soft_delete: bool = True
 ) -> None:
@@ -838,7 +801,7 @@ async def add_store_agent_to_library(
                 "isCreatedByUser": False,
                 "useGraphIsActiveVersion": False,
                 "settings": SafeJson(
-                    _initialize_graph_settings(graph_model).model_dump()
+                    GraphSettings.from_graph(graph_model).model_dump()
                 ),
             },
             include=library_agent_include(
@@ -1228,8 +1191,15 @@ async def fork_library_agent(
         )
         new_graph = await on_graph_activate(new_graph, user_id=user_id)
 
-        # Create a library agent for the new graph
-        return (await create_library_agent(new_graph, user_id))[0]
+        # Create a library agent for the new graph, preserving safe mode settings
+        return (
+            await create_library_agent(
+                new_graph,
+                user_id,
+                hitl_safe_mode=original_agent.settings.human_in_the_loop_safe_mode,
+                sensitive_action_safe_mode=original_agent.settings.sensitive_action_safe_mode,
+            )
+        )[0]
     except prisma.errors.PrismaError as e:
         logger.error(f"Database error cloning library agent: {e}")
         raise DatabaseError("Failed to fork library agent") from e

autogpt_platform/backend/backend/api/features/library/model.py

@@ -73,6 +73,12 @@ class LibraryAgent(pydantic.BaseModel):
     has_external_trigger: bool = pydantic.Field(
         description="Whether the agent has an external trigger (e.g. webhook) node"
     )
+    has_human_in_the_loop: bool = pydantic.Field(
+        description="Whether the agent has human-in-the-loop blocks"
+    )
+    has_sensitive_action: bool = pydantic.Field(
+        description="Whether the agent has sensitive action blocks"
+    )
     trigger_setup_info: Optional[GraphTriggerInfo] = None
 
     # Indicates whether there's a new output (based on recent runs)
@@ -180,6 +186,8 @@ class LibraryAgent(pydantic.BaseModel):
                 graph.credentials_input_schema if sub_graphs is not None else None
             ),
             has_external_trigger=graph.has_external_trigger,
+            has_human_in_the_loop=graph.has_human_in_the_loop,
+            has_sensitive_action=graph.has_sensitive_action,
             trigger_setup_info=graph.trigger_setup_info,
             new_output=new_output,
             can_access_graph=can_access_graph,

autogpt_platform/backend/backend/api/features/library/routes_test.py

@@ -52,6 +52,8 @@ async def test_get_library_agents_success(
                 output_schema={"type": "object", "properties": {}},
                 credentials_input_schema={"type": "object", "properties": {}},
                 has_external_trigger=False,
+                has_human_in_the_loop=False,
+                has_sensitive_action=False,
                 status=library_model.LibraryAgentStatus.COMPLETED,
                 recommended_schedule_cron=None,
                 new_output=False,
@@ -75,6 +77,8 @@ async def test_get_library_agents_success(
                 output_schema={"type": "object", "properties": {}},
                 credentials_input_schema={"type": "object", "properties": {}},
                 has_external_trigger=False,
+                has_human_in_the_loop=False,
+                has_sensitive_action=False,
                 status=library_model.LibraryAgentStatus.COMPLETED,
                 recommended_schedule_cron=None,
                 new_output=False,
@@ -150,6 +154,8 @@ async def test_get_favorite_library_agents_success(
                 output_schema={"type": "object", "properties": {}},
                 credentials_input_schema={"type": "object", "properties": {}},
                 has_external_trigger=False,
+                has_human_in_the_loop=False,
+                has_sensitive_action=False,
                 status=library_model.LibraryAgentStatus.COMPLETED,
                 recommended_schedule_cron=None,
                 new_output=False,
@@ -218,6 +224,8 @@ def test_add_agent_to_library_success(
         output_schema={"type": "object", "properties": {}},
         credentials_input_schema={"type": "object", "properties": {}},
         has_external_trigger=False,
+        has_human_in_the_loop=False,
+        has_sensitive_action=False,
         status=library_model.LibraryAgentStatus.COMPLETED,
         new_output=False,
         can_access_graph=True,

autogpt_platform/backend/backend/api/features/store/content_handlers.py

@@ -225,28 +225,6 @@ class BlockHandler(ContentHandler):
                     [cat.value for cat in categories] if categories else []
                 )
 
-                # Extract provider names from credentials fields
-                provider_names: list[str] = []
-                is_integration = False
-                if hasattr(block_instance, "input_schema"):
-                    credentials_info = (
-                        block_instance.input_schema.get_credentials_fields_info()
-                    )
-                    is_integration = len(credentials_info) > 0
-                    for info in credentials_info.values():
-                        for provider in info.provider:
-                            provider_names.append(provider.value.lower())
-
-                # Check if block has LlmModel field in input schema
-                has_llm_model_field = False
-                if hasattr(block_instance, "input_schema"):
-                    from backend.blocks.llm import LlmModel
-
-                    for field in block_instance.input_schema.model_fields.values():
-                        if field.annotation == LlmModel:
-                            has_llm_model_field = True
-                            break
-
                 items.append(
                     ContentItem(
                         content_id=block_id,
@@ -255,9 +233,6 @@ class BlockHandler(ContentHandler):
                         metadata={
                             "name": getattr(block_instance, "name", ""),
                             "categories": categories_list,
-                            "providers": provider_names,
-                            "has_llm_model_field": has_llm_model_field,
-                            "is_integration": is_integration,
                         },
                         user_id=None,  # Blocks are public
                     )

autogpt_platform/backend/backend/api/features/store/embeddings.py

@@ -154,6 +154,7 @@ async def store_content_embedding(
 
         # Upsert the embedding
         # WHERE clause in DO UPDATE prevents PostgreSQL 15 bug with NULLS NOT DISTINCT
+        # Use unqualified ::vector - pgvector is in search_path on all environments
         await execute_raw_with_schema(
             """
             INSERT INTO {schema_prefix}"UnifiedContentEmbedding" (
@@ -177,7 +178,6 @@ async def store_content_embedding(
             searchable_text,
             metadata_json,
             client=client,
-            set_public_search_path=True,
         )
 
         logger.info(f"Stored embedding for {content_type}:{content_id}")
@@ -236,7 +236,6 @@ async def get_content_embedding(
             content_type,
             content_id,
             user_id,
-            set_public_search_path=True,
         )
 
         if result and len(result) > 0:
@@ -871,31 +870,45 @@ async def semantic_search(
         # Add content type parameters and build placeholders dynamically
         content_type_start_idx = len(params) + 1
         content_type_placeholders = ", ".join(
-            f'${content_type_start_idx + i}::{{{{schema_prefix}}}}"ContentType"'
+            "$" + str(content_type_start_idx + i) + '::{schema_prefix}"ContentType"'
             for i in range(len(content_types))
         )
         params.extend([ct.value for ct in content_types])
 
-        sql = f"""
+        # Build min_similarity param index before appending
+        min_similarity_idx = len(params) + 1
+        params.append(min_similarity)
+
+        # Use unqualified ::vector and <=> operator - pgvector is in search_path on all environments
+        sql = (
+            """
             SELECT
                 "contentId" as content_id,
                 "contentType" as content_type,
                 "searchableText" as searchable_text,
                 metadata,
-                1 - (embedding <=> '{embedding_str}'::vector) as similarity
-            FROM {{{{schema_prefix}}}}"UnifiedContentEmbedding"
-            WHERE "contentType" IN ({content_type_placeholders})
-            {user_filter}
-            AND 1 - (embedding <=> '{embedding_str}'::vector) >= ${len(params) + 1}
+                1 - (embedding <=> '"""
+            + embedding_str
+            + """'::vector) as similarity
+            FROM {schema_prefix}"UnifiedContentEmbedding"
+            WHERE "contentType" IN ("""
+            + content_type_placeholders
+            + """)
+            """
+            + user_filter
+            + """
+            AND 1 - (embedding <=> '"""
+            + embedding_str
+            + """'::vector) >= $"""
+            + str(min_similarity_idx)
+            + """
             ORDER BY similarity DESC
             LIMIT $1
         """
-        params.append(min_similarity)
+        )
 
         try:
-            results = await query_raw_with_schema(
-                sql, *params, set_public_search_path=True
-            )
+            results = await query_raw_with_schema(sql, *params)
             return [
                 {
                     "content_id": row["content_id"],
@@ -922,31 +935,41 @@ async def semantic_search(
     # Add content type parameters and build placeholders dynamically
     content_type_start_idx = len(params_lexical) + 1
     content_type_placeholders_lexical = ", ".join(
-        f'${content_type_start_idx + i}::{{{{schema_prefix}}}}"ContentType"'
+        "$" + str(content_type_start_idx + i) + '::{schema_prefix}"ContentType"'
         for i in range(len(content_types))
     )
     params_lexical.extend([ct.value for ct in content_types])
 
-    sql_lexical = f"""
+    # Build query param index before appending
+    query_param_idx = len(params_lexical) + 1
+    params_lexical.append(f"%{query}%")
+
+    # Use regular string (not f-string) for template to preserve {schema_prefix} placeholders
+    sql_lexical = (
+        """
         SELECT
             "contentId" as content_id,
             "contentType" as content_type,
             "searchableText" as searchable_text,
             metadata,
             0.0 as similarity
-        FROM {{{{schema_prefix}}}}"UnifiedContentEmbedding"
-        WHERE "contentType" IN ({content_type_placeholders_lexical})
-        {user_filter}
-        AND "searchableText" ILIKE ${len(params_lexical) + 1}
+        FROM {schema_prefix}"UnifiedContentEmbedding"
+        WHERE "contentType" IN ("""
+        + content_type_placeholders_lexical
+        + """)
+        """
+        + user_filter
+        + """
+        AND "searchableText" ILIKE $"""
+        + str(query_param_idx)
+        + """
         ORDER BY "updatedAt" DESC
         LIMIT $1
     """
-    params_lexical.append(f"%{query}%")
+    )
 
     try:
-        results = await query_raw_with_schema(
-            sql_lexical, *params_lexical, set_public_search_path=True
-        )
+        results = await query_raw_with_schema(sql_lexical, *params_lexical)
         return [
             {
                 "content_id": row["content_id"],

autogpt_platform/backend/backend/api/features/store/embeddings_test.py

@@ -155,18 +155,14 @@ async def test_store_embedding_success(mocker):
     )
 
     assert result is True
-    # execute_raw is called twice: once for SET search_path, once for INSERT
-    assert mock_client.execute_raw.call_count == 2
+    # execute_raw is called once for INSERT (no separate SET search_path needed)
+    assert mock_client.execute_raw.call_count == 1
 
-    # First call: SET search_path
-    first_call_args = mock_client.execute_raw.call_args_list[0][0]
-    assert "SET search_path" in first_call_args[0]
-
-    # Second call: INSERT query with the actual data
-    second_call_args = mock_client.execute_raw.call_args_list[1][0]
-    assert "test-version-id" in second_call_args
-    assert "[0.1,0.2,0.3]" in second_call_args
-    assert None in second_call_args  # userId should be None for store agents
+    # Verify the INSERT query with the actual data
+    call_args = mock_client.execute_raw.call_args_list[0][0]
+    assert "test-version-id" in call_args
+    assert "[0.1,0.2,0.3]" in call_args
+    assert None in call_args  # userId should be None for store agents
 
 
 @pytest.mark.asyncio(loop_scope="session")

autogpt_platform/backend/backend/api/features/store/hybrid_search.py

@@ -12,7 +12,7 @@ from dataclasses import dataclass
 from typing import Any, Literal
 
 from prisma.enums import ContentType
-from rank_bm25 import BM25Okapi
+from rank_bm25 import BM25Okapi  # type: ignore[import-untyped]
 
 from backend.api.features.store.embeddings import (
     EMBEDDING_DIM,
@@ -363,9 +363,7 @@ async def unified_hybrid_search(
         LIMIT {limit_param} OFFSET {offset_param}
     """
 
-    results = await query_raw_with_schema(
-        sql_query, *params, set_public_search_path=True
-    )
+    results = await query_raw_with_schema(sql_query, *params)
 
     total = results[0]["total_count"] if results else 0
     # Apply BM25 reranking
@@ -688,9 +686,7 @@ async def hybrid_search(
         LIMIT {limit_param} OFFSET {offset_param}
     """
 
-    results = await query_raw_with_schema(
-        sql_query, *params, set_public_search_path=True
-    )
+    results = await query_raw_with_schema(sql_query, *params)
 
     total = results[0]["total_count"] if results else 0
 

autogpt_platform/backend/backend/api/features/v1.py

@@ -761,10 +761,8 @@ async def create_new_graph(
     graph.reassign_ids(user_id=user_id, reassign_graph_id=True)
     graph.validate_graph(for_run=False)
 
-    # The return value of the create graph & library function is intentionally not used here,
-    # as the graph already valid and no sub-graphs are returned back.
     await graph_db.create_graph(graph, user_id=user_id)
-    await library_db.create_library_agent(graph, user_id=user_id)
+    await library_db.create_library_agent(graph, user_id)
     activated_graph = await on_graph_activate(graph, user_id=user_id)
 
     if create_graph.source == "builder":
@@ -888,21 +886,19 @@ async def set_graph_active_version(
 async def _update_library_agent_version_and_settings(
     user_id: str, agent_graph: graph_db.GraphModel
 ) -> library_model.LibraryAgent:
-    # Keep the library agent up to date with the new active version
     library = await library_db.update_agent_version_in_library(
         user_id, agent_graph.id, agent_graph.version
     )
-    # If the graph has HITL node, initialize the setting if it's not already set.
-    if (
-        agent_graph.has_human_in_the_loop
-        and library.settings.human_in_the_loop_safe_mode is None
-    ):
-        await library_db.update_library_agent_settings(
+    updated_settings = GraphSettings.from_graph(
+        graph=agent_graph,
+        hitl_safe_mode=library.settings.human_in_the_loop_safe_mode,
+        sensitive_action_safe_mode=library.settings.sensitive_action_safe_mode,
+    )
+    if updated_settings != library.settings:
+        library = await library_db.update_library_agent(
+            library_agent_id=library.id,
             user_id=user_id,
-            agent_id=library.id,
-            settings=library.settings.model_copy(
-                update={"human_in_the_loop_safe_mode": True}
-            ),
+            settings=updated_settings,
         )
     return library
 
@@ -919,21 +915,18 @@ async def update_graph_settings(
     user_id: Annotated[str, Security(get_user_id)],
 ) -> GraphSettings:
     """Update graph settings for the user's library agent."""
-    # Get the library agent for this graph
     library_agent = await library_db.get_library_agent_by_graph_id(
         graph_id=graph_id, user_id=user_id
     )
     if not library_agent:
         raise HTTPException(404, f"Graph #{graph_id} not found in user's library")
 
-    # Update the library agent settings
-    updated_agent = await library_db.update_library_agent_settings(
+    updated_agent = await library_db.update_library_agent(
+        library_agent_id=library_agent.id,
         user_id=user_id,
-        agent_id=library_agent.id,
         settings=settings,
     )
 
-    # Return the updated settings
     return GraphSettings.model_validate(updated_agent.settings)
 
 

autogpt_platform/backend/backend/blocks/data_manipulation.py

@@ -680,3 +680,58 @@ class ListIsEmptyBlock(Block):
 
     async def run(self, input_data: Input, **kwargs) -> BlockOutput:
         yield "is_empty", len(input_data.list) == 0
+
+
+class ConcatenateListsBlock(Block):
+    class Input(BlockSchemaInput):
+        lists: List[List[Any]] = SchemaField(
+            description="A list of lists to concatenate together. All lists will be combined in order into a single list.",
+            placeholder="e.g., [[1, 2], [3, 4], [5, 6]]",
+        )
+
+    class Output(BlockSchemaOutput):
+        concatenated_list: List[Any] = SchemaField(
+            description="The concatenated list containing all elements from all input lists in order."
+        )
+        error: str = SchemaField(
+            description="Error message if concatenation failed due to invalid input types."
+        )
+
+    def __init__(self):
+        super().__init__(
+            id="3cf9298b-5817-4141-9d80-7c2cc5199c8e",
+            description="Concatenates multiple lists into a single list. All elements from all input lists are combined in order.",
+            categories={BlockCategory.BASIC},
+            input_schema=ConcatenateListsBlock.Input,
+            output_schema=ConcatenateListsBlock.Output,
+            test_input=[
+                {"lists": [[1, 2, 3], [4, 5, 6]]},
+                {"lists": [["a", "b"], ["c"], ["d", "e", "f"]]},
+                {"lists": [[1, 2], []]},
+                {"lists": []},
+            ],
+            test_output=[
+                ("concatenated_list", [1, 2, 3, 4, 5, 6]),
+                ("concatenated_list", ["a", "b", "c", "d", "e", "f"]),
+                ("concatenated_list", [1, 2]),
+                ("concatenated_list", []),
+            ],
+        )
+
+    async def run(self, input_data: Input, **kwargs) -> BlockOutput:
+        concatenated = []
+        for idx, lst in enumerate(input_data.lists):
+            if lst is None:
+                # Skip None values to avoid errors
+                continue
+            if not isinstance(lst, list):
+                # Type validation: each item must be a list
+                # Strings are iterable and would cause extend() to iterate character-by-character
+                # Non-iterable types would raise TypeError
+                yield "error", (
+                    f"Invalid input at index {idx}: expected a list, got {type(lst).__name__}. "
+                    f"All items in 'lists' must be lists (e.g., [[1, 2], [3, 4]])."
+                )
+                return
+            concatenated.extend(lst)
+        yield "concatenated_list", concatenated

autogpt_platform/backend/backend/blocks/helpers/review.py

@@ -84,7 +84,7 @@ class HITLReviewHelper:
             Exception: If review creation or status update fails
         """
         # Skip review if safe mode is disabled - return auto-approved result
-        if not execution_context.safe_mode:
+        if not execution_context.human_in_the_loop_safe_mode:
             logger.info(
                 f"Block {block_name} skipping review for node {node_exec_id} - safe mode disabled"
             )

autogpt_platform/backend/backend/blocks/human_in_the_loop.py

@@ -104,7 +104,7 @@ class HumanInTheLoopBlock(Block):
         execution_context: ExecutionContext,
         **_kwargs,
     ) -> BlockOutput:
-        if not execution_context.safe_mode:
+        if not execution_context.human_in_the_loop_safe_mode:
             logger.info(
                 f"HITL block skipping review for node {node_exec_id} - safe mode disabled"
             )

autogpt_platform/backend/backend/blocks/llm.py

@@ -79,6 +79,10 @@ class ModelMetadata(NamedTuple):
     provider: str
     context_window: int
     max_output_tokens: int | None
+    display_name: str
+    provider_name: str
+    creator_name: str
+    price_tier: Literal[1, 2, 3]
 
 
 class LlmModelMeta(EnumMeta):
@@ -171,6 +175,26 @@ class LlmModel(str, Enum, metaclass=LlmModelMeta):
     V0_1_5_LG = "v0-1.5-lg"
     V0_1_0_MD = "v0-1.0-md"
 
+    @classmethod
+    def __get_pydantic_json_schema__(cls, schema, handler):
+        json_schema = handler(schema)
+        llm_model_metadata = {}
+        for model in cls:
+            model_name = model.value
+            metadata = model.metadata
+            llm_model_metadata[model_name] = {
+                "creator": metadata.creator_name,
+                "creator_name": metadata.creator_name,
+                "title": metadata.display_name,
+                "provider": metadata.provider,
+                "provider_name": metadata.provider_name,
+                "name": model_name,
+                "price_tier": metadata.price_tier,
+            }
+        json_schema["llm_model"] = True
+        json_schema["llm_model_metadata"] = llm_model_metadata
+        return json_schema
+
     @property
     def metadata(self) -> ModelMetadata:
         return MODEL_METADATA[self]
@@ -190,119 +214,291 @@ class LlmModel(str, Enum, metaclass=LlmModelMeta):
 
 MODEL_METADATA = {
     # https://platform.openai.com/docs/models
-    LlmModel.O3: ModelMetadata("openai", 200000, 100000),
-    LlmModel.O3_MINI: ModelMetadata("openai", 200000, 100000),  # o3-mini-2025-01-31
-    LlmModel.O1: ModelMetadata("openai", 200000, 100000),  # o1-2024-12-17
-    LlmModel.O1_MINI: ModelMetadata("openai", 128000, 65536),  # o1-mini-2024-09-12
+    LlmModel.O3: ModelMetadata("openai", 200000, 100000, "O3", "OpenAI", "OpenAI", 2),
+    LlmModel.O3_MINI: ModelMetadata(
+        "openai", 200000, 100000, "O3 Mini", "OpenAI", "OpenAI", 1
+    ),  # o3-mini-2025-01-31
+    LlmModel.O1: ModelMetadata(
+        "openai", 200000, 100000, "O1", "OpenAI", "OpenAI", 3
+    ),  # o1-2024-12-17
+    LlmModel.O1_MINI: ModelMetadata(
+        "openai", 128000, 65536, "O1 Mini", "OpenAI", "OpenAI", 2
+    ),  # o1-mini-2024-09-12
     # GPT-5 models
-    LlmModel.GPT5_2: ModelMetadata("openai", 400000, 128000),
-    LlmModel.GPT5_1: ModelMetadata("openai", 400000, 128000),
-    LlmModel.GPT5: ModelMetadata("openai", 400000, 128000),
-    LlmModel.GPT5_MINI: ModelMetadata("openai", 400000, 128000),
-    LlmModel.GPT5_NANO: ModelMetadata("openai", 400000, 128000),
-    LlmModel.GPT5_CHAT: ModelMetadata("openai", 400000, 16384),
-    LlmModel.GPT41: ModelMetadata("openai", 1047576, 32768),
-    LlmModel.GPT41_MINI: ModelMetadata("openai", 1047576, 32768),
+    LlmModel.GPT5_2: ModelMetadata(
+        "openai", 400000, 128000, "GPT-5.2", "OpenAI", "OpenAI", 3
+    ),
+    LlmModel.GPT5_1: ModelMetadata(
+        "openai", 400000, 128000, "GPT-5.1", "OpenAI", "OpenAI", 2
+    ),
+    LlmModel.GPT5: ModelMetadata(
+        "openai", 400000, 128000, "GPT-5", "OpenAI", "OpenAI", 1
+    ),
+    LlmModel.GPT5_MINI: ModelMetadata(
+        "openai", 400000, 128000, "GPT-5 Mini", "OpenAI", "OpenAI", 1
+    ),
+    LlmModel.GPT5_NANO: ModelMetadata(
+        "openai", 400000, 128000, "GPT-5 Nano", "OpenAI", "OpenAI", 1
+    ),
+    LlmModel.GPT5_CHAT: ModelMetadata(
+        "openai", 400000, 16384, "GPT-5 Chat Latest", "OpenAI", "OpenAI", 2
+    ),
+    LlmModel.GPT41: ModelMetadata(
+        "openai", 1047576, 32768, "GPT-4.1", "OpenAI", "OpenAI", 1
+    ),
+    LlmModel.GPT41_MINI: ModelMetadata(
+        "openai", 1047576, 32768, "GPT-4.1 Mini", "OpenAI", "OpenAI", 1
+    ),
     LlmModel.GPT4O_MINI: ModelMetadata(
-        "openai", 128000, 16384
+        "openai", 128000, 16384, "GPT-4o Mini", "OpenAI", "OpenAI", 1
     ),  # gpt-4o-mini-2024-07-18
-    LlmModel.GPT4O: ModelMetadata("openai", 128000, 16384),  # gpt-4o-2024-08-06
+    LlmModel.GPT4O: ModelMetadata(
+        "openai", 128000, 16384, "GPT-4o", "OpenAI", "OpenAI", 2
+    ),  # gpt-4o-2024-08-06
     LlmModel.GPT4_TURBO: ModelMetadata(
-        "openai", 128000, 4096
+        "openai", 128000, 4096, "GPT-4 Turbo", "OpenAI", "OpenAI", 3
     ),  # gpt-4-turbo-2024-04-09
-    LlmModel.GPT3_5_TURBO: ModelMetadata("openai", 16385, 4096),  # gpt-3.5-turbo-0125
+    LlmModel.GPT3_5_TURBO: ModelMetadata(
+        "openai", 16385, 4096, "GPT-3.5 Turbo", "OpenAI", "OpenAI", 1
+    ),  # gpt-3.5-turbo-0125
     # https://docs.anthropic.com/en/docs/about-claude/models
     LlmModel.CLAUDE_4_1_OPUS: ModelMetadata(
-        "anthropic", 200000, 32000
+        "anthropic", 200000, 32000, "Claude Opus 4.1", "Anthropic", "Anthropic", 3
     ),  # claude-opus-4-1-20250805
     LlmModel.CLAUDE_4_OPUS: ModelMetadata(
-        "anthropic", 200000, 32000
+        "anthropic", 200000, 32000, "Claude Opus 4", "Anthropic", "Anthropic", 3
     ),  # claude-4-opus-20250514
     LlmModel.CLAUDE_4_SONNET: ModelMetadata(
-        "anthropic", 200000, 64000
+        "anthropic", 200000, 64000, "Claude Sonnet 4", "Anthropic", "Anthropic", 2
     ),  # claude-4-sonnet-20250514
     LlmModel.CLAUDE_4_5_OPUS: ModelMetadata(
-        "anthropic", 200000, 64000
+        "anthropic", 200000, 64000, "Claude Opus 4.5", "Anthropic", "Anthropic", 3
     ),  # claude-opus-4-5-20251101
     LlmModel.CLAUDE_4_5_SONNET: ModelMetadata(
-        "anthropic", 200000, 64000
+        "anthropic", 200000, 64000, "Claude Sonnet 4.5", "Anthropic", "Anthropic", 3
     ),  # claude-sonnet-4-5-20250929
     LlmModel.CLAUDE_4_5_HAIKU: ModelMetadata(
-        "anthropic", 200000, 64000
+        "anthropic", 200000, 64000, "Claude Haiku 4.5", "Anthropic", "Anthropic", 2
     ),  # claude-haiku-4-5-20251001
     LlmModel.CLAUDE_3_7_SONNET: ModelMetadata(
-        "anthropic", 200000, 64000
+        "anthropic", 200000, 64000, "Claude 3.7 Sonnet", "Anthropic", "Anthropic", 2
     ),  # claude-3-7-sonnet-20250219
     LlmModel.CLAUDE_3_HAIKU: ModelMetadata(
-        "anthropic", 200000, 4096
+        "anthropic", 200000, 4096, "Claude 3 Haiku", "Anthropic", "Anthropic", 1
     ),  # claude-3-haiku-20240307
     # https://docs.aimlapi.com/api-overview/model-database/text-models
-    LlmModel.AIML_API_QWEN2_5_72B: ModelMetadata("aiml_api", 32000, 8000),
-    LlmModel.AIML_API_LLAMA3_1_70B: ModelMetadata("aiml_api", 128000, 40000),
-    LlmModel.AIML_API_LLAMA3_3_70B: ModelMetadata("aiml_api", 128000, None),
-    LlmModel.AIML_API_META_LLAMA_3_1_70B: ModelMetadata("aiml_api", 131000, 2000),
-    LlmModel.AIML_API_LLAMA_3_2_3B: ModelMetadata("aiml_api", 128000, None),
-    # https://console.groq.com/docs/models
-    LlmModel.LLAMA3_3_70B: ModelMetadata("groq", 128000, 32768),
-    LlmModel.LLAMA3_1_8B: ModelMetadata("groq", 128000, 8192),
-    # https://ollama.com/library
-    LlmModel.OLLAMA_LLAMA3_3: ModelMetadata("ollama", 8192, None),
-    LlmModel.OLLAMA_LLAMA3_2: ModelMetadata("ollama", 8192, None),
-    LlmModel.OLLAMA_LLAMA3_8B: ModelMetadata("ollama", 8192, None),
-    LlmModel.OLLAMA_LLAMA3_405B: ModelMetadata("ollama", 8192, None),
-    LlmModel.OLLAMA_DOLPHIN: ModelMetadata("ollama", 32768, None),
-    # https://openrouter.ai/models
-    LlmModel.GEMINI_2_5_PRO: ModelMetadata("open_router", 1050000, 8192),
-    LlmModel.GEMINI_3_PRO_PREVIEW: ModelMetadata("open_router", 1048576, 65535),
-    LlmModel.GEMINI_2_5_FLASH: ModelMetadata("open_router", 1048576, 65535),
-    LlmModel.GEMINI_2_0_FLASH: ModelMetadata("open_router", 1048576, 8192),
-    LlmModel.GEMINI_2_5_FLASH_LITE_PREVIEW: ModelMetadata(
-        "open_router", 1048576, 65535
+    LlmModel.AIML_API_QWEN2_5_72B: ModelMetadata(
+        "aiml_api", 32000, 8000, "Qwen 2.5 72B Instruct Turbo", "AI/ML", "Qwen", 1
+    ),
+    LlmModel.AIML_API_LLAMA3_1_70B: ModelMetadata(
+        "aiml_api",
+        128000,
+        40000,
+        "Llama 3.1 Nemotron 70B Instruct",
+        "AI/ML",
+        "Nvidia",
+        1,
+    ),
+    LlmModel.AIML_API_LLAMA3_3_70B: ModelMetadata(
+        "aiml_api", 128000, None, "Llama 3.3 70B Instruct Turbo", "AI/ML", "Meta", 1
+    ),
+    LlmModel.AIML_API_META_LLAMA_3_1_70B: ModelMetadata(
+        "aiml_api", 131000, 2000, "Llama 3.1 70B Instruct Turbo", "AI/ML", "Meta", 1
+    ),
+    LlmModel.AIML_API_LLAMA_3_2_3B: ModelMetadata(
+        "aiml_api", 128000, None, "Llama 3.2 3B Instruct Turbo", "AI/ML", "Meta", 1
+    ),
+    # https://console.groq.com/docs/models
+    LlmModel.LLAMA3_3_70B: ModelMetadata(
+        "groq", 128000, 32768, "Llama 3.3 70B Versatile", "Groq", "Meta", 1
+    ),
+    LlmModel.LLAMA3_1_8B: ModelMetadata(
+        "groq", 128000, 8192, "Llama 3.1 8B Instant", "Groq", "Meta", 1
+    ),
+    # https://ollama.com/library
+    LlmModel.OLLAMA_LLAMA3_3: ModelMetadata(
+        "ollama", 8192, None, "Llama 3.3", "Ollama", "Meta", 1
+    ),
+    LlmModel.OLLAMA_LLAMA3_2: ModelMetadata(
+        "ollama", 8192, None, "Llama 3.2", "Ollama", "Meta", 1
+    ),
+    LlmModel.OLLAMA_LLAMA3_8B: ModelMetadata(
+        "ollama", 8192, None, "Llama 3", "Ollama", "Meta", 1
+    ),
+    LlmModel.OLLAMA_LLAMA3_405B: ModelMetadata(
+        "ollama", 8192, None, "Llama 3.1 405B", "Ollama", "Meta", 1
+    ),
+    LlmModel.OLLAMA_DOLPHIN: ModelMetadata(
+        "ollama", 32768, None, "Dolphin Mistral Latest", "Ollama", "Mistral AI", 1
+    ),
+    # https://openrouter.ai/models
+    LlmModel.GEMINI_2_5_PRO: ModelMetadata(
+        "open_router",
+        1050000,
+        8192,
+        "Gemini 2.5 Pro Preview 03.25",
+        "OpenRouter",
+        "Google",
+        2,
+    ),
+    LlmModel.GEMINI_3_PRO_PREVIEW: ModelMetadata(
+        "open_router", 1048576, 65535, "Gemini 3 Pro Preview", "OpenRouter", "Google", 2
+    ),
+    LlmModel.GEMINI_2_5_FLASH: ModelMetadata(
+        "open_router", 1048576, 65535, "Gemini 2.5 Flash", "OpenRouter", "Google", 1
+    ),
+    LlmModel.GEMINI_2_0_FLASH: ModelMetadata(
+        "open_router", 1048576, 8192, "Gemini 2.0 Flash 001", "OpenRouter", "Google", 1
+    ),
+    LlmModel.GEMINI_2_5_FLASH_LITE_PREVIEW: ModelMetadata(
+        "open_router",
+        1048576,
+        65535,
+        "Gemini 2.5 Flash Lite Preview 06.17",
+        "OpenRouter",
+        "Google",
+        1,
+    ),
+    LlmModel.GEMINI_2_0_FLASH_LITE: ModelMetadata(
+        "open_router",
+        1048576,
+        8192,
+        "Gemini 2.0 Flash Lite 001",
+        "OpenRouter",
+        "Google",
+        1,
+    ),
+    LlmModel.MISTRAL_NEMO: ModelMetadata(
+        "open_router", 128000, 4096, "Mistral Nemo", "OpenRouter", "Mistral AI", 1
+    ),
+    LlmModel.COHERE_COMMAND_R_08_2024: ModelMetadata(
+        "open_router", 128000, 4096, "Command R 08.2024", "OpenRouter", "Cohere", 1
+    ),
+    LlmModel.COHERE_COMMAND_R_PLUS_08_2024: ModelMetadata(
+        "open_router", 128000, 4096, "Command R Plus 08.2024", "OpenRouter", "Cohere", 2
+    ),
+    LlmModel.DEEPSEEK_CHAT: ModelMetadata(
+        "open_router", 64000, 2048, "DeepSeek Chat", "OpenRouter", "DeepSeek", 1
+    ),
+    LlmModel.DEEPSEEK_R1_0528: ModelMetadata(
+        "open_router", 163840, 163840, "DeepSeek R1 0528", "OpenRouter", "DeepSeek", 1
+    ),
+    LlmModel.PERPLEXITY_SONAR: ModelMetadata(
+        "open_router", 127000, 8000, "Sonar", "OpenRouter", "Perplexity", 1
+    ),
+    LlmModel.PERPLEXITY_SONAR_PRO: ModelMetadata(
+        "open_router", 200000, 8000, "Sonar Pro", "OpenRouter", "Perplexity", 2
     ),
-    LlmModel.GEMINI_2_0_FLASH_LITE: ModelMetadata("open_router", 1048576, 8192),
-    LlmModel.MISTRAL_NEMO: ModelMetadata("open_router", 128000, 4096),
-    LlmModel.COHERE_COMMAND_R_08_2024: ModelMetadata("open_router", 128000, 4096),
-    LlmModel.COHERE_COMMAND_R_PLUS_08_2024: ModelMetadata("open_router", 128000, 4096),
-    LlmModel.DEEPSEEK_CHAT: ModelMetadata("open_router", 64000, 2048),
-    LlmModel.DEEPSEEK_R1_0528: ModelMetadata("open_router", 163840, 163840),
-    LlmModel.PERPLEXITY_SONAR: ModelMetadata("open_router", 127000, 8000),
-    LlmModel.PERPLEXITY_SONAR_PRO: ModelMetadata("open_router", 200000, 8000),
     LlmModel.PERPLEXITY_SONAR_DEEP_RESEARCH: ModelMetadata(
         "open_router",
         128000,
         16000,
+        "Sonar Deep Research",
+        "OpenRouter",
+        "Perplexity",
+        3,
     ),
     LlmModel.NOUSRESEARCH_HERMES_3_LLAMA_3_1_405B: ModelMetadata(
-        "open_router", 131000, 4096
+        "open_router",
+        131000,
+        4096,
+        "Hermes 3 Llama 3.1 405B",
+        "OpenRouter",
+        "Nous Research",
+        1,
     ),
     LlmModel.NOUSRESEARCH_HERMES_3_LLAMA_3_1_70B: ModelMetadata(
-        "open_router", 12288, 12288
+        "open_router",
+        12288,
+        12288,
+        "Hermes 3 Llama 3.1 70B",
+        "OpenRouter",
+        "Nous Research",
+        1,
+    ),
+    LlmModel.OPENAI_GPT_OSS_120B: ModelMetadata(
+        "open_router", 131072, 131072, "GPT-OSS 120B", "OpenRouter", "OpenAI", 1
+    ),
+    LlmModel.OPENAI_GPT_OSS_20B: ModelMetadata(
+        "open_router", 131072, 32768, "GPT-OSS 20B", "OpenRouter", "OpenAI", 1
+    ),
+    LlmModel.AMAZON_NOVA_LITE_V1: ModelMetadata(
+        "open_router", 300000, 5120, "Nova Lite V1", "OpenRouter", "Amazon", 1
+    ),
+    LlmModel.AMAZON_NOVA_MICRO_V1: ModelMetadata(
+        "open_router", 128000, 5120, "Nova Micro V1", "OpenRouter", "Amazon", 1
+    ),
+    LlmModel.AMAZON_NOVA_PRO_V1: ModelMetadata(
+        "open_router", 300000, 5120, "Nova Pro V1", "OpenRouter", "Amazon", 1
+    ),
+    LlmModel.MICROSOFT_WIZARDLM_2_8X22B: ModelMetadata(
+        "open_router", 65536, 4096, "WizardLM 2 8x22B", "OpenRouter", "Microsoft", 1
+    ),
+    LlmModel.GRYPHE_MYTHOMAX_L2_13B: ModelMetadata(
+        "open_router", 4096, 4096, "MythoMax L2 13B", "OpenRouter", "Gryphe", 1
+    ),
+    LlmModel.META_LLAMA_4_SCOUT: ModelMetadata(
+        "open_router", 131072, 131072, "Llama 4 Scout", "OpenRouter", "Meta", 1
+    ),
+    LlmModel.META_LLAMA_4_MAVERICK: ModelMetadata(
+        "open_router", 1048576, 1000000, "Llama 4 Maverick", "OpenRouter", "Meta", 1
+    ),
+    LlmModel.GROK_4: ModelMetadata(
+        "open_router", 256000, 256000, "Grok 4", "OpenRouter", "xAI", 3
+    ),
+    LlmModel.GROK_4_FAST: ModelMetadata(
+        "open_router", 2000000, 30000, "Grok 4 Fast", "OpenRouter", "xAI", 1
+    ),
+    LlmModel.GROK_4_1_FAST: ModelMetadata(
+        "open_router", 2000000, 30000, "Grok 4.1 Fast", "OpenRouter", "xAI", 1
+    ),
+    LlmModel.GROK_CODE_FAST_1: ModelMetadata(
+        "open_router", 256000, 10000, "Grok Code Fast 1", "OpenRouter", "xAI", 1
+    ),
+    LlmModel.KIMI_K2: ModelMetadata(
+        "open_router", 131000, 131000, "Kimi K2", "OpenRouter", "Moonshot AI", 1
+    ),
+    LlmModel.QWEN3_235B_A22B_THINKING: ModelMetadata(
+        "open_router",
+        262144,
+        262144,
+        "Qwen 3 235B A22B Thinking 2507",
+        "OpenRouter",
+        "Qwen",
+        1,
+    ),
+    LlmModel.QWEN3_CODER: ModelMetadata(
+        "open_router", 262144, 262144, "Qwen 3 Coder", "OpenRouter", "Qwen", 3
     ),
-    LlmModel.OPENAI_GPT_OSS_120B: ModelMetadata("open_router", 131072, 131072),
-    LlmModel.OPENAI_GPT_OSS_20B: ModelMetadata("open_router", 131072, 32768),
-    LlmModel.AMAZON_NOVA_LITE_V1: ModelMetadata("open_router", 300000, 5120),
-    LlmModel.AMAZON_NOVA_MICRO_V1: ModelMetadata("open_router", 128000, 5120),
-    LlmModel.AMAZON_NOVA_PRO_V1: ModelMetadata("open_router", 300000, 5120),
-    LlmModel.MICROSOFT_WIZARDLM_2_8X22B: ModelMetadata("open_router", 65536, 4096),
-    LlmModel.GRYPHE_MYTHOMAX_L2_13B: ModelMetadata("open_router", 4096, 4096),
-    LlmModel.META_LLAMA_4_SCOUT: ModelMetadata("open_router", 131072, 131072),
-    LlmModel.META_LLAMA_4_MAVERICK: ModelMetadata("open_router", 1048576, 1000000),
-    LlmModel.GROK_4: ModelMetadata("open_router", 256000, 256000),
-    LlmModel.GROK_4_FAST: ModelMetadata("open_router", 2000000, 30000),
-    LlmModel.GROK_4_1_FAST: ModelMetadata("open_router", 2000000, 30000),
-    LlmModel.GROK_CODE_FAST_1: ModelMetadata("open_router", 256000, 10000),
-    LlmModel.KIMI_K2: ModelMetadata("open_router", 131000, 131000),
-    LlmModel.QWEN3_235B_A22B_THINKING: ModelMetadata("open_router", 262144, 262144),
-    LlmModel.QWEN3_CODER: ModelMetadata("open_router", 262144, 262144),
     # Llama API models
-    LlmModel.LLAMA_API_LLAMA_4_SCOUT: ModelMetadata("llama_api", 128000, 4028),
-    LlmModel.LLAMA_API_LLAMA4_MAVERICK: ModelMetadata("llama_api", 128000, 4028),
-    LlmModel.LLAMA_API_LLAMA3_3_8B: ModelMetadata("llama_api", 128000, 4028),
-    LlmModel.LLAMA_API_LLAMA3_3_70B: ModelMetadata("llama_api", 128000, 4028),
+    LlmModel.LLAMA_API_LLAMA_4_SCOUT: ModelMetadata(
+        "llama_api",
+        128000,
+        4028,
+        "Llama 4 Scout 17B 16E Instruct FP8",
+        "Llama API",
+        "Meta",
+        1,
+    ),
+    LlmModel.LLAMA_API_LLAMA4_MAVERICK: ModelMetadata(
+        "llama_api",
+        128000,
+        4028,
+        "Llama 4 Maverick 17B 128E Instruct FP8",
+        "Llama API",
+        "Meta",
+        1,
+    ),
+    LlmModel.LLAMA_API_LLAMA3_3_8B: ModelMetadata(
+        "llama_api", 128000, 4028, "Llama 3.3 8B Instruct", "Llama API", "Meta", 1
+    ),
+    LlmModel.LLAMA_API_LLAMA3_3_70B: ModelMetadata(
+        "llama_api", 128000, 4028, "Llama 3.3 70B Instruct", "Llama API", "Meta", 1
+    ),
     # v0 by Vercel models
-    LlmModel.V0_1_5_MD: ModelMetadata("v0", 128000, 64000),
-    LlmModel.V0_1_5_LG: ModelMetadata("v0", 512000, 64000),
-    LlmModel.V0_1_0_MD: ModelMetadata("v0", 128000, 64000),
+    LlmModel.V0_1_5_MD: ModelMetadata("v0", 128000, 64000, "v0 1.5 MD", "V0", "V0", 1),
+    LlmModel.V0_1_5_LG: ModelMetadata("v0", 512000, 64000, "v0 1.5 LG", "V0", "V0", 1),
+    LlmModel.V0_1_0_MD: ModelMetadata("v0", 128000, 64000, "v0 1.0 MD", "V0", "V0", 1),
 }
 
 DEFAULT_LLM_MODEL = LlmModel.GPT5_2

autogpt_platform/backend/backend/blocks/test/test_smart_decision_maker.py

@@ -242,7 +242,7 @@ async def test_smart_decision_maker_tracks_llm_stats():
         outputs = {}
         # Create execution context
 
-        mock_execution_context = ExecutionContext(safe_mode=False)
+        mock_execution_context = ExecutionContext(human_in_the_loop_safe_mode=False)
 
         # Create a mock execution processor for tests
 
@@ -343,7 +343,7 @@ async def test_smart_decision_maker_parameter_validation():
 
         # Create execution context
 
-        mock_execution_context = ExecutionContext(safe_mode=False)
+        mock_execution_context = ExecutionContext(human_in_the_loop_safe_mode=False)
 
         # Create a mock execution processor for tests
 
@@ -409,7 +409,7 @@ async def test_smart_decision_maker_parameter_validation():
 
         # Create execution context
 
-        mock_execution_context = ExecutionContext(safe_mode=False)
+        mock_execution_context = ExecutionContext(human_in_the_loop_safe_mode=False)
 
         # Create a mock execution processor for tests
 
@@ -471,7 +471,7 @@ async def test_smart_decision_maker_parameter_validation():
         outputs = {}
         # Create execution context
 
-        mock_execution_context = ExecutionContext(safe_mode=False)
+        mock_execution_context = ExecutionContext(human_in_the_loop_safe_mode=False)
 
         # Create a mock execution processor for tests
 
@@ -535,7 +535,7 @@ async def test_smart_decision_maker_parameter_validation():
         outputs = {}
         # Create execution context
 
-        mock_execution_context = ExecutionContext(safe_mode=False)
+        mock_execution_context = ExecutionContext(human_in_the_loop_safe_mode=False)
 
         # Create a mock execution processor for tests
 
@@ -658,7 +658,7 @@ async def test_smart_decision_maker_raw_response_conversion():
         outputs = {}
         # Create execution context
 
-        mock_execution_context = ExecutionContext(safe_mode=False)
+        mock_execution_context = ExecutionContext(human_in_the_loop_safe_mode=False)
 
         # Create a mock execution processor for tests
 
@@ -730,7 +730,7 @@ async def test_smart_decision_maker_raw_response_conversion():
         outputs = {}
         # Create execution context
 
-        mock_execution_context = ExecutionContext(safe_mode=False)
+        mock_execution_context = ExecutionContext(human_in_the_loop_safe_mode=False)
 
         # Create a mock execution processor for tests
 
@@ -786,7 +786,7 @@ async def test_smart_decision_maker_raw_response_conversion():
         outputs = {}
         # Create execution context
 
-        mock_execution_context = ExecutionContext(safe_mode=False)
+        mock_execution_context = ExecutionContext(human_in_the_loop_safe_mode=False)
 
         # Create a mock execution processor for tests
 
@@ -905,7 +905,7 @@ async def test_smart_decision_maker_agent_mode():
         # Create a mock execution context
 
         mock_execution_context = ExecutionContext(
-            safe_mode=False,
+            human_in_the_loop_safe_mode=False,
         )
 
         # Create a mock execution processor for agent mode tests
@@ -1027,7 +1027,7 @@ async def test_smart_decision_maker_traditional_mode_default():
 
         # Create execution context
 
-        mock_execution_context = ExecutionContext(safe_mode=False)
+        mock_execution_context = ExecutionContext(human_in_the_loop_safe_mode=False)
 
         # Create a mock execution processor for tests
 

autogpt_platform/backend/backend/blocks/test/test_smart_decision_maker_dynamic_fields.py

@@ -386,7 +386,7 @@ async def test_output_yielding_with_dynamic_fields():
             outputs = {}
             from backend.data.execution import ExecutionContext
 
-            mock_execution_context = ExecutionContext(safe_mode=False)
+            mock_execution_context = ExecutionContext(human_in_the_loop_safe_mode=False)
             mock_execution_processor = MagicMock()
 
             async for output_name, output_value in block.run(
@@ -609,7 +609,9 @@ async def test_validation_errors_dont_pollute_conversation():
                 outputs = {}
                 from backend.data.execution import ExecutionContext
 
-                mock_execution_context = ExecutionContext(safe_mode=False)
+                mock_execution_context = ExecutionContext(
+                    human_in_the_loop_safe_mode=False
+                )
 
                 # Create a proper mock execution processor for agent mode
                 from collections import defaultdict

autogpt_platform/backend/backend/data/block.py

@@ -474,7 +474,7 @@ class Block(ABC, Generic[BlockSchemaInputType, BlockSchemaOutputType]):
         self.block_type = block_type
         self.webhook_config = webhook_config
         self.execution_stats: NodeExecutionStats = NodeExecutionStats()
-        self.requires_human_review: bool = False
+        self.is_sensitive_action: bool = False
 
         if self.webhook_config:
             if isinstance(self.webhook_config, BlockWebhookConfig):
@@ -637,8 +637,9 @@ class Block(ABC, Generic[BlockSchemaInputType, BlockSchemaOutputType]):
             - should_pause: True if execution should be paused for review
             - input_data_to_use: The input data to use (may be modified by reviewer)
         """
-        # Skip review if not required or safe mode is disabled
-        if not self.requires_human_review or not execution_context.safe_mode:
+        if not (
+            self.is_sensitive_action and execution_context.sensitive_action_safe_mode
+        ):
             return False, input_data
 
         from backend.blocks.helpers.review import HITLReviewHelper

autogpt_platform/backend/backend/data/block_cost_config.py

@@ -99,10 +99,15 @@ MODEL_COST: dict[LlmModel, int] = {
     LlmModel.OPENAI_GPT_OSS_20B: 1,
     LlmModel.GEMINI_2_5_PRO: 4,
     LlmModel.GEMINI_3_PRO_PREVIEW: 5,
+    LlmModel.GEMINI_2_5_FLASH: 1,
+    LlmModel.GEMINI_2_0_FLASH: 1,
+    LlmModel.GEMINI_2_5_FLASH_LITE_PREVIEW: 1,
+    LlmModel.GEMINI_2_0_FLASH_LITE: 1,
     LlmModel.MISTRAL_NEMO: 1,
     LlmModel.COHERE_COMMAND_R_08_2024: 1,
     LlmModel.COHERE_COMMAND_R_PLUS_08_2024: 3,
     LlmModel.DEEPSEEK_CHAT: 2,
+    LlmModel.DEEPSEEK_R1_0528: 1,
     LlmModel.PERPLEXITY_SONAR: 1,
     LlmModel.PERPLEXITY_SONAR_PRO: 5,
     LlmModel.PERPLEXITY_SONAR_DEEP_RESEARCH: 10,
@@ -126,11 +131,6 @@ MODEL_COST: dict[LlmModel, int] = {
     LlmModel.KIMI_K2: 1,
     LlmModel.QWEN3_235B_A22B_THINKING: 1,
     LlmModel.QWEN3_CODER: 9,
-    LlmModel.GEMINI_2_5_FLASH: 1,
-    LlmModel.GEMINI_2_0_FLASH: 1,
-    LlmModel.GEMINI_2_5_FLASH_LITE_PREVIEW: 1,
-    LlmModel.GEMINI_2_0_FLASH_LITE: 1,
-    LlmModel.DEEPSEEK_R1_0528: 1,
     # v0 by Vercel models
     LlmModel.V0_1_5_MD: 1,
     LlmModel.V0_1_5_LG: 2,

autogpt_platform/backend/backend/data/db.py

@@ -38,20 +38,6 @@ POOL_TIMEOUT = os.getenv("DB_POOL_TIMEOUT")
 if POOL_TIMEOUT:
     DATABASE_URL = add_param(DATABASE_URL, "pool_timeout", POOL_TIMEOUT)
 
-# Add public schema to search_path for pgvector type access
-# The vector extension is in public schema, but search_path is determined by schema parameter
-# Extract the schema from DATABASE_URL or default to 'public' (matching get_database_schema())
-parsed_url = urlparse(DATABASE_URL)
-url_params = dict(parse_qsl(parsed_url.query))
-db_schema = url_params.get("schema", "public")
-# Build search_path, avoiding duplicates if db_schema is already 'public'
-search_path_schemas = list(
-    dict.fromkeys([db_schema, "public"])
-)  # Preserves order, removes duplicates
-search_path = ",".join(search_path_schemas)
-# This allows using ::vector without schema qualification
-DATABASE_URL = add_param(DATABASE_URL, "options", f"-c search_path={search_path}")
-
 HTTP_TIMEOUT = int(POOL_TIMEOUT) if POOL_TIMEOUT else None
 
 prisma = Prisma(
@@ -127,38 +113,48 @@ async def _raw_with_schema(
     *args,
     execute: bool = False,
     client: Prisma | None = None,
-    set_public_search_path: bool = False,
 ) -> list[dict] | int:
     """Internal: Execute raw SQL with proper schema handling.
 
     Use query_raw_with_schema() or execute_raw_with_schema() instead.
 
+    Supports placeholders:
+        - {schema_prefix}: Table/type prefix (e.g., "platform".)
+        - {schema}: Raw schema name for application tables (e.g., platform)
+
+    Note on pgvector types:
+        Use unqualified ::vector and <=> operator in queries. PostgreSQL resolves
+        these via search_path, which includes the schema where pgvector is installed
+        on all environments (local, CI, dev).
+
     Args:
-        query_template: SQL query with {schema_prefix} placeholder
+        query_template: SQL query with {schema_prefix} and/or {schema} placeholders
         *args: Query parameters
         execute: If False, executes SELECT query. If True, executes INSERT/UPDATE/DELETE.
         client: Optional Prisma client for transactions (only used when execute=True).
-        set_public_search_path: If True, sets search_path to include public schema.
-                                Needed for pgvector types and other public schema objects.
 
     Returns:
         - list[dict] if execute=False (query results)
         - int if execute=True (number of affected rows)
+
+    Example with vector type:
+        await execute_raw_with_schema(
+            'INSERT INTO {schema_prefix}"Embedding" (vec) VALUES ($1::vector)',
+            embedding_data
+        )
     """
     schema = get_database_schema()
     schema_prefix = f'"{schema}".' if schema != "public" else ""
-    formatted_query = query_template.format(schema_prefix=schema_prefix)
+
+    formatted_query = query_template.format(
+        schema_prefix=schema_prefix,
+        schema=schema,
+    )
 
     import prisma as prisma_module
 
     db_client = client if client else prisma_module.get_client()
 
-    # Set search_path to include public schema if requested
-    # Prisma doesn't support the 'options' connection parameter, so we set it per-session
-    # This is idempotent and safe to call multiple times
-    if set_public_search_path:
-        await db_client.execute_raw(f"SET search_path = {schema}, public")  # type: ignore
-
     if execute:
         result = await db_client.execute_raw(formatted_query, *args)  # type: ignore
     else:
@@ -167,16 +163,12 @@ async def _raw_with_schema(
     return result
 
 
-async def query_raw_with_schema(
-    query_template: str, *args, set_public_search_path: bool = False
-) -> list[dict]:
+async def query_raw_with_schema(query_template: str, *args) -> list[dict]:
     """Execute raw SQL SELECT query with proper schema handling.
 
     Args:
-        query_template: SQL query with {schema_prefix} placeholder
+        query_template: SQL query with {schema_prefix} and/or {schema} placeholders
         *args: Query parameters
-        set_public_search_path: If True, sets search_path to include public schema.
-                                Needed for pgvector types and other public schema objects.
 
     Returns:
         List of result rows as dictionaries
@@ -187,23 +179,20 @@ async def query_raw_with_schema(
             user_id
         )
     """
-    return await _raw_with_schema(query_template, *args, execute=False, set_public_search_path=set_public_search_path)  # type: ignore
+    return await _raw_with_schema(query_template, *args, execute=False)  # type: ignore
 
 
 async def execute_raw_with_schema(
     query_template: str,
     *args,
     client: Prisma | None = None,
-    set_public_search_path: bool = False,
 ) -> int:
     """Execute raw SQL command (INSERT/UPDATE/DELETE) with proper schema handling.
 
     Args:
-        query_template: SQL query with {schema_prefix} placeholder
+        query_template: SQL query with {schema_prefix} and/or {schema} placeholders
         *args: Query parameters
         client: Optional Prisma client for transactions
-        set_public_search_path: If True, sets search_path to include public schema.
-                                Needed for pgvector types and other public schema objects.
 
     Returns:
         Number of affected rows
@@ -215,7 +204,7 @@ async def execute_raw_with_schema(
             client=tx  # Optional transaction client
         )
     """
-    return await _raw_with_schema(query_template, *args, execute=True, client=client, set_public_search_path=set_public_search_path)  # type: ignore
+    return await _raw_with_schema(query_template, *args, execute=True, client=client)  # type: ignore
 
 
 class BaseDbModel(BaseModel):

autogpt_platform/backend/backend/data/event_bus.py

@@ -103,8 +103,18 @@ class RedisEventBus(BaseRedisEventBus[M], ABC):
         return redis.get_redis()
 
     def publish_event(self, event: M, channel_key: str):
-        message, full_channel_name = self._serialize_message(event, channel_key)
-        self.connection.publish(full_channel_name, message)
+        """
+        Publish an event to Redis. Gracefully handles connection failures
+        by logging the error instead of raising exceptions.
+        """
+        try:
+            message, full_channel_name = self._serialize_message(event, channel_key)
+            self.connection.publish(full_channel_name, message)
+        except Exception:
+            logger.exception(
+                f"Failed to publish event to Redis channel {channel_key}. "
+                "Event bus operation will continue without Redis connectivity."
+            )
 
     def listen_events(self, channel_key: str) -> Generator[M, None, None]:
         pubsub, full_channel_name = self._get_pubsub_channel(
@@ -128,9 +138,19 @@ class AsyncRedisEventBus(BaseRedisEventBus[M], ABC):
         return await redis.get_redis_async()
 
     async def publish_event(self, event: M, channel_key: str):
-        message, full_channel_name = self._serialize_message(event, channel_key)
-        connection = await self.connection
-        await connection.publish(full_channel_name, message)
+        """
+        Publish an event to Redis. Gracefully handles connection failures
+        by logging the error instead of raising exceptions.
+        """
+        try:
+            message, full_channel_name = self._serialize_message(event, channel_key)
+            connection = await self.connection
+            await connection.publish(full_channel_name, message)
+        except Exception:
+            logger.exception(
+                f"Failed to publish event to Redis channel {channel_key}. "
+                "Event bus operation will continue without Redis connectivity."
+            )
 
     async def listen_events(self, channel_key: str) -> AsyncGenerator[M, None]:
         pubsub, full_channel_name = self._get_pubsub_channel(

autogpt_platform/backend/backend/data/event_bus_test.py

@@ -0,0 +1,56 @@
+"""
+Tests for event_bus graceful degradation when Redis is unavailable.
+"""
+
+from unittest.mock import AsyncMock, patch
+
+import pytest
+from pydantic import BaseModel
+
+from backend.data.event_bus import AsyncRedisEventBus
+
+
+class TestEvent(BaseModel):
+    """Test event model."""
+
+    message: str
+
+
+class TestNotificationBus(AsyncRedisEventBus[TestEvent]):
+    """Test implementation of AsyncRedisEventBus."""
+
+    Model = TestEvent
+
+    @property
+    def event_bus_name(self) -> str:
+        return "test_event_bus"
+
+
+@pytest.mark.asyncio
+async def test_publish_event_handles_connection_failure_gracefully():
+    """Test that publish_event logs exception instead of raising when Redis is unavailable."""
+    bus = TestNotificationBus()
+    event = TestEvent(message="test message")
+
+    # Mock get_redis_async to raise connection error
+    with patch(
+        "backend.data.event_bus.redis.get_redis_async",
+        side_effect=ConnectionError("Authentication required."),
+    ):
+        # Should not raise exception
+        await bus.publish_event(event, "test_channel")
+
+
+@pytest.mark.asyncio
+async def test_publish_event_works_with_redis_available():
+    """Test that publish_event works normally when Redis is available."""
+    bus = TestNotificationBus()
+    event = TestEvent(message="test message")
+
+    # Mock successful Redis connection
+    mock_redis = AsyncMock()
+    mock_redis.publish = AsyncMock()
+
+    with patch("backend.data.event_bus.redis.get_redis_async", return_value=mock_redis):
+        await bus.publish_event(event, "test_channel")
+        mock_redis.publish.assert_called_once()

autogpt_platform/backend/backend/data/execution.py

@@ -81,7 +81,10 @@ class ExecutionContext(BaseModel):
     This includes information needed by blocks, sub-graphs, and execution management.
     """
 
-    safe_mode: bool = True
+    model_config = {"extra": "ignore"}
+
+    human_in_the_loop_safe_mode: bool = True
+    sensitive_action_safe_mode: bool = False
     user_timezone: str = "UTC"
     root_execution_id: Optional[str] = None
     parent_execution_id: Optional[str] = None

autogpt_platform/backend/backend/data/graph.py

@@ -3,7 +3,7 @@ import logging
 import uuid
 from collections import defaultdict
 from datetime import datetime, timezone
-from typing import TYPE_CHECKING, Any, Literal, Optional, cast
+from typing import TYPE_CHECKING, Annotated, Any, Literal, Optional, cast
 
 from prisma.enums import SubmissionStatus
 from prisma.models import (
@@ -20,7 +20,7 @@ from prisma.types import (
     AgentNodeLinkCreateInput,
     StoreListingVersionWhereInput,
 )
-from pydantic import BaseModel, Field, create_model
+from pydantic import BaseModel, BeforeValidator, Field, create_model
 from pydantic.fields import computed_field
 
 from backend.blocks.agent import AgentExecutorBlock
@@ -62,7 +62,31 @@ logger = logging.getLogger(__name__)
 
 
 class GraphSettings(BaseModel):
-    human_in_the_loop_safe_mode: bool | None = None
+    # Use Annotated with BeforeValidator to coerce None to default values.
+    # This handles cases where the database has null values for these fields.
+    model_config = {"extra": "ignore"}
+
+    human_in_the_loop_safe_mode: Annotated[
+        bool, BeforeValidator(lambda v: v if v is not None else True)
+    ] = True
+    sensitive_action_safe_mode: Annotated[
+        bool, BeforeValidator(lambda v: v if v is not None else False)
+    ] = False
+
+    @classmethod
+    def from_graph(
+        cls,
+        graph: "GraphModel",
+        hitl_safe_mode: bool | None = None,
+        sensitive_action_safe_mode: bool = False,
+    ) -> "GraphSettings":
+        # Default to True if not explicitly set
+        if hitl_safe_mode is None:
+            hitl_safe_mode = True
+        return cls(
+            human_in_the_loop_safe_mode=hitl_safe_mode,
+            sensitive_action_safe_mode=sensitive_action_safe_mode,
+        )
 
 
 class Link(BaseDbModel):
@@ -244,10 +268,14 @@ class BaseGraph(BaseDbModel):
         return any(
             node.block_id
             for node in self.nodes
-            if (
-                node.block.block_type == BlockType.HUMAN_IN_THE_LOOP
-                or node.block.requires_human_review
-            )
+            if node.block.block_type == BlockType.HUMAN_IN_THE_LOOP
+        )
+
+    @computed_field
+    @property
+    def has_sensitive_action(self) -> bool:
+        return any(
+            node.block_id for node in self.nodes if node.block.is_sensitive_action
         )
 
     @property

autogpt_platform/backend/backend/data/understanding.py

@@ -328,6 +328,8 @@ async def clear_business_understanding(user_id: str) -> bool:
 
 def format_understanding_for_prompt(understanding: BusinessUnderstanding) -> str:
     """Format business understanding as text for system prompt injection."""
+    if not understanding:
+        return ""
     sections = []
 
     # User info section

autogpt_platform/backend/backend/executor/scheduler.py

@@ -309,7 +309,7 @@ def ensure_embeddings_coverage():
 
         # Process in batches until no more missing embeddings
         while True:
-            result = db_client.backfill_missing_embeddings(batch_size=10)
+            result = db_client.backfill_missing_embeddings(batch_size=100)
 
             total_processed += result["processed"]
             total_success += result["success"]

autogpt_platform/backend/backend/executor/utils.py

@@ -873,11 +873,8 @@ async def add_graph_execution(
         settings = await gdb.get_graph_settings(user_id=user_id, graph_id=graph_id)
 
         execution_context = ExecutionContext(
-            safe_mode=(
-                settings.human_in_the_loop_safe_mode
-                if settings.human_in_the_loop_safe_mode is not None
-                else True
-            ),
+            human_in_the_loop_safe_mode=settings.human_in_the_loop_safe_mode,
+            sensitive_action_safe_mode=settings.sensitive_action_safe_mode,
             user_timezone=(
                 user.timezone if user.timezone != USER_TIMEZONE_NOT_SET else "UTC"
             ),

autogpt_platform/backend/backend/executor/utils_test.py

@@ -386,6 +386,7 @@ async def test_add_graph_execution_is_repeatable(mocker: MockerFixture):
     mock_user.timezone = "UTC"
     mock_settings = mocker.MagicMock()
     mock_settings.human_in_the_loop_safe_mode = True
+    mock_settings.sensitive_action_safe_mode = False
 
     mock_udb.get_user_by_id = mocker.AsyncMock(return_value=mock_user)
     mock_gdb.get_graph_settings = mocker.AsyncMock(return_value=mock_settings)
@@ -651,6 +652,7 @@ async def test_add_graph_execution_with_nodes_to_skip(mocker: MockerFixture):
     mock_user.timezone = "UTC"
     mock_settings = mocker.MagicMock()
     mock_settings.human_in_the_loop_safe_mode = True
+    mock_settings.sensitive_action_safe_mode = False
 
     mock_udb.get_user_by_id = mocker.AsyncMock(return_value=mock_user)
     mock_gdb.get_graph_settings = mocker.AsyncMock(return_value=mock_settings)

autogpt_platform/backend/migrations/20250120120000_add_suggested_blocks_materialized_view/migration.sql

@@ -1,102 +0,0 @@
--- This migration creates a materialized view for suggested blocks based on execution counts
--- The view aggregates execution counts per block for the last 14 days
---
--- IMPORTANT: For production environments, pg_cron is REQUIRED for automatic refresh
--- Prerequisites for production:
---   1. pg_cron extension must be installed: CREATE EXTENSION pg_cron;
---   2. pg_cron must be configured in postgresql.conf:
---      shared_preload_libraries = 'pg_cron'
---      cron.database_name = 'your_database_name'
---
--- For development environments without pg_cron:
---   The migration will succeed but you must manually refresh views with:
---   SELECT refresh_suggested_blocks_view();
-
--- Check if pg_cron extension is installed and set a flag
-DO $$
-DECLARE
-    has_pg_cron BOOLEAN;
-BEGIN
-    SELECT EXISTS (SELECT 1 FROM pg_extension WHERE extname = 'pg_cron') INTO has_pg_cron;
-
-    IF NOT has_pg_cron THEN
-        RAISE WARNING 'pg_cron extension is not installed!';
-        RAISE WARNING 'Materialized view will be created but WILL NOT refresh automatically.';
-        RAISE WARNING 'For production use, install pg_cron with: CREATE EXTENSION pg_cron;';
-        RAISE WARNING 'For development, manually refresh with: SELECT refresh_suggested_blocks_view();';
-    END IF;
-
-    -- Store the flag for later use in the migration
-    PERFORM set_config('migration.has_pg_cron', has_pg_cron::text, false);
-END
-$$;
-
--- Create materialized view for suggested blocks based on execution counts in last 14 days
--- The 14-day threshold is hardcoded to ensure consistent behavior
-CREATE MATERIALIZED VIEW IF NOT EXISTS "mv_suggested_blocks" AS
-SELECT
-    agent_node."agentBlockId" AS block_id,
-    COUNT(execution.id) AS execution_count
-FROM "AgentNodeExecution" execution
-JOIN "AgentNode" agent_node ON execution."agentNodeId" = agent_node.id
-WHERE execution."endedTime" >= (NOW() - INTERVAL '14 days')
-GROUP BY agent_node."agentBlockId"
-ORDER BY execution_count DESC;
-
--- Create unique index for concurrent refresh support
-CREATE UNIQUE INDEX IF NOT EXISTS "idx_mv_suggested_blocks_block_id" ON "mv_suggested_blocks"("block_id");
-
--- Create refresh function
-CREATE OR REPLACE FUNCTION refresh_suggested_blocks_view()
-RETURNS void
-LANGUAGE plpgsql
-AS $$
-DECLARE
-    target_schema text := current_schema();
-BEGIN
-    -- Use CONCURRENTLY for better performance during refresh
-    REFRESH MATERIALIZED VIEW CONCURRENTLY "mv_suggested_blocks";
-    RAISE NOTICE 'Suggested blocks materialized view refreshed in schema % at %', target_schema, NOW();
-EXCEPTION
-    WHEN OTHERS THEN
-        -- Fallback to non-concurrent refresh if concurrent fails
-        REFRESH MATERIALIZED VIEW "mv_suggested_blocks";
-        RAISE NOTICE 'Suggested blocks materialized view refreshed (non-concurrent) in schema % at %. Concurrent refresh failed due to: %', target_schema, NOW(), SQLERRM;
-END;
-$$;
-
--- Schedule automatic refresh every hour (only if pg_cron is available)
-DO $$
-DECLARE
-    has_pg_cron BOOLEAN;
-    current_schema_name text := current_schema();
-    old_job_name text;
-    job_name text;
-BEGIN
-    -- Check if pg_cron extension exists
-    SELECT EXISTS (SELECT 1 FROM pg_extension WHERE extname = 'pg_cron') INTO has_pg_cron;
-
-    IF has_pg_cron THEN
-        job_name := format('refresh-suggested-blocks_%s', current_schema_name);
-
-        -- Try to unschedule existing job (ignore errors if it doesn't exist)
-        BEGIN
-            PERFORM cron.unschedule(job_name);
-        EXCEPTION WHEN OTHERS THEN
-            NULL;
-        END;
-
-        -- Schedule the new job to run every hour
-        PERFORM cron.schedule(
-            job_name,
-            '0 * * * *',  -- Every hour at minute 0
-            format('SET search_path TO %I; SELECT refresh_suggested_blocks_view();', current_schema_name)
-        );
-        RAISE NOTICE 'Scheduled job %; runs every hour for schema %', job_name, current_schema_name;
-    ELSE
-        RAISE WARNING 'Automatic refresh NOT configured - pg_cron is not available';
-        RAISE WARNING 'You must manually refresh the view with: SELECT refresh_suggested_blocks_view();';
-        RAISE WARNING 'Or install pg_cron for automatic refresh in production';
-    END IF;
-END;
-$$;

autogpt_platform/backend/migrations/20260109181714_add_docs_embedding/migration.sql

@@ -1,11 +1,37 @@
 -- CreateExtension
 -- Supabase: pgvector must be enabled via Dashboard → Database → Extensions first
--- Create in public schema so vector type is available across all schemas
+-- Ensures vector extension is in the current schema (from DATABASE_URL ?schema= param)
+-- If it exists in a different schema (e.g., public), we drop and recreate it in the current schema
+-- This ensures vector type is in the same schema as tables, making ::vector work without explicit qualification
 DO $$
+DECLARE
+    current_schema_name text;
+    vector_schema text;
 BEGIN
-    CREATE EXTENSION IF NOT EXISTS "vector" WITH SCHEMA "public";
-EXCEPTION WHEN OTHERS THEN
-    RAISE NOTICE 'vector extension not available or already exists, skipping';
+    -- Get the current schema from search_path
+    SELECT current_schema() INTO current_schema_name;
+
+    -- Check if vector extension exists and which schema it's in
+    SELECT n.nspname INTO vector_schema
+    FROM pg_extension e
+    JOIN pg_namespace n ON e.extnamespace = n.oid
+    WHERE e.extname = 'vector';
+
+    -- Handle removal if in wrong schema
+    IF vector_schema IS NOT NULL AND vector_schema != current_schema_name THEN
+        BEGIN
+            -- Vector exists in a different schema, drop it first
+            RAISE WARNING 'pgvector found in schema "%" but need it in "%". Dropping and reinstalling...',
+                vector_schema, current_schema_name;
+            EXECUTE 'DROP EXTENSION IF EXISTS vector CASCADE';
+        EXCEPTION WHEN OTHERS THEN
+            RAISE EXCEPTION 'Failed to drop pgvector from schema "%": %. You may need to drop it manually.',
+                vector_schema, SQLERRM;
+        END;
+    END IF;
+
+    -- Create extension in current schema (let it fail naturally if not available)
+    EXECUTE format('CREATE EXTENSION IF NOT EXISTS vector SCHEMA %I', current_schema_name);
 END $$;
 
 -- CreateEnum
@@ -19,7 +45,7 @@ CREATE TABLE "UnifiedContentEmbedding" (
     "contentType" "ContentType" NOT NULL,
     "contentId" TEXT NOT NULL,
     "userId" TEXT,
-    "embedding" public.vector(1536) NOT NULL,
+    "embedding" vector(1536) NOT NULL,
     "searchableText" TEXT NOT NULL,
     "metadata" JSONB NOT NULL DEFAULT '{}',
 
@@ -45,4 +71,4 @@ CREATE UNIQUE INDEX "UnifiedContentEmbedding_contentType_contentId_userId_key" O
 -- Uses cosine distance operator (<=>), which matches the query in hybrid_search.py
 -- Note: Drop first in case Prisma created a btree index (Prisma doesn't support HNSW)
 DROP INDEX IF EXISTS "UnifiedContentEmbedding_embedding_idx";
-CREATE INDEX "UnifiedContentEmbedding_embedding_idx" ON "UnifiedContentEmbedding" USING hnsw ("embedding" public.vector_cosine_ops);
+CREATE INDEX "UnifiedContentEmbedding_embedding_idx" ON "UnifiedContentEmbedding" USING hnsw ("embedding" vector_cosine_ops);

autogpt_platform/backend/migrations/20260112173500_add_supabase_extensions_to_platform_schema/migration.sql

@@ -1,71 +0,0 @@
--- Acknowledge Supabase-managed extensions to prevent drift warnings
--- These extensions are pre-installed by Supabase in specific schemas
--- This migration ensures they exist where available (Supabase) or skips gracefully (CI)
-
--- Create schemas (safe in both CI and Supabase)
-CREATE SCHEMA IF NOT EXISTS "extensions";
-
--- Extensions that exist in both CI and Supabase
-DO $$
-BEGIN
-    CREATE EXTENSION IF NOT EXISTS "pgcrypto" WITH SCHEMA "extensions";
-EXCEPTION WHEN OTHERS THEN
-    RAISE NOTICE 'pgcrypto extension not available, skipping';
-END $$;
-
-DO $$
-BEGIN
-    CREATE EXTENSION IF NOT EXISTS "uuid-ossp" WITH SCHEMA "extensions";
-EXCEPTION WHEN OTHERS THEN
-    RAISE NOTICE 'uuid-ossp extension not available, skipping';
-END $$;
-
--- Supabase-specific extensions (skip gracefully in CI)
-DO $$
-BEGIN
-    CREATE EXTENSION IF NOT EXISTS "pg_stat_statements" WITH SCHEMA "extensions";
-EXCEPTION WHEN OTHERS THEN
-    RAISE NOTICE 'pg_stat_statements extension not available, skipping';
-END $$;
-
-DO $$
-BEGIN
-    CREATE EXTENSION IF NOT EXISTS "pg_net" WITH SCHEMA "extensions";
-EXCEPTION WHEN OTHERS THEN
-    RAISE NOTICE 'pg_net extension not available, skipping';
-END $$;
-
-DO $$
-BEGIN
-    CREATE EXTENSION IF NOT EXISTS "pgjwt" WITH SCHEMA "extensions";
-EXCEPTION WHEN OTHERS THEN
-    RAISE NOTICE 'pgjwt extension not available, skipping';
-END $$;
-
-DO $$
-BEGIN
-    CREATE SCHEMA IF NOT EXISTS "graphql";
-    CREATE EXTENSION IF NOT EXISTS "pg_graphql" WITH SCHEMA "graphql";
-EXCEPTION WHEN OTHERS THEN
-    RAISE NOTICE 'pg_graphql extension not available, skipping';
-END $$;
-
-DO $$
-BEGIN
-    CREATE SCHEMA IF NOT EXISTS "pgsodium";
-    CREATE EXTENSION IF NOT EXISTS "pgsodium" WITH SCHEMA "pgsodium";
-EXCEPTION WHEN OTHERS THEN
-    RAISE NOTICE 'pgsodium extension not available, skipping';
-END $$;
-
-DO $$
-BEGIN
-    CREATE SCHEMA IF NOT EXISTS "vault";
-    CREATE EXTENSION IF NOT EXISTS "supabase_vault" WITH SCHEMA "vault";
-EXCEPTION WHEN OTHERS THEN
-    RAISE NOTICE 'supabase_vault extension not available, skipping';
-END $$;
-
-
--- Return to platform
-CREATE SCHEMA IF NOT EXISTS "platform";

autogpt_platform/backend/scripts/generate_block_docs.py

@@ -366,12 +366,12 @@ def generate_block_markdown(
     lines.append("")
 
     # What it is (full description)
-    lines.append(f"### What it is")
+    lines.append("### What it is")
     lines.append(block.description or "No description available.")
     lines.append("")
 
     # How it works (manual section)
-    lines.append(f"### How it works")
+    lines.append("### How it works")
     how_it_works = manual_content.get(
         "how_it_works", "_Add technical explanation here._"
     )
@@ -383,7 +383,7 @@ def generate_block_markdown(
     # Inputs table (auto-generated)
     visible_inputs = [f for f in block.inputs if not f.hidden]
     if visible_inputs:
-        lines.append(f"### Inputs")
+        lines.append("### Inputs")
         lines.append("")
         lines.append("| Input | Description | Type | Required |")
         lines.append("|-------|-------------|------|----------|")
@@ -400,7 +400,7 @@ def generate_block_markdown(
     # Outputs table (auto-generated)
     visible_outputs = [f for f in block.outputs if not f.hidden]
     if visible_outputs:
-        lines.append(f"### Outputs")
+        lines.append("### Outputs")
         lines.append("")
         lines.append("| Output | Description | Type |")
         lines.append("|--------|-------------|------|")
@@ -414,7 +414,7 @@ def generate_block_markdown(
         lines.append("")
 
     # Possible use case (manual section)
-    lines.append(f"### Possible use case")
+    lines.append("### Possible use case")
     use_case = manual_content.get("use_case", "_Add practical use case examples here._")
     lines.append("<!-- MANUAL: use_case -->")
     lines.append(use_case)

autogpt_platform/backend/snapshots/grph_single

@@ -11,6 +11,7 @@
   "forked_from_version": null,
   "has_external_trigger": false,
   "has_human_in_the_loop": false,
+  "has_sensitive_action": false,
   "id": "graph-123",
   "input_schema": {
     "properties": {},

autogpt_platform/backend/snapshots/grphs_all

@@ -11,6 +11,7 @@
     "forked_from_version": null,
     "has_external_trigger": false,
     "has_human_in_the_loop": false,
+    "has_sensitive_action": false,
     "id": "graph-123",
     "input_schema": {
       "properties": {},

autogpt_platform/backend/snapshots/lib_agts_search

@@ -27,6 +27,8 @@
         "properties": {}
       },
       "has_external_trigger": false,
+      "has_human_in_the_loop": false,
+      "has_sensitive_action": false,
       "trigger_setup_info": null,
       "new_output": false,
       "can_access_graph": true,
@@ -34,7 +36,8 @@
       "is_favorite": false,
       "recommended_schedule_cron": null,
       "settings": {
-        "human_in_the_loop_safe_mode": null
+        "human_in_the_loop_safe_mode": true,
+        "sensitive_action_safe_mode": false
       },
       "marketplace_listing": null
     },
@@ -65,6 +68,8 @@
         "properties": {}
       },
       "has_external_trigger": false,
+      "has_human_in_the_loop": false,
+      "has_sensitive_action": false,
       "trigger_setup_info": null,
       "new_output": false,
       "can_access_graph": false,
@@ -72,7 +77,8 @@
       "is_favorite": false,
       "recommended_schedule_cron": null,
       "settings": {
-        "human_in_the_loop_safe_mode": null
+        "human_in_the_loop_safe_mode": true,
+        "sensitive_action_safe_mode": false
       },
       "marketplace_listing": null
     }

autogpt_platform/frontend/.env.default

@@ -29,4 +29,4 @@ NEXT_PUBLIC_CLOUDFLARE_TURNSTILE_SITE_KEY=
 NEXT_PUBLIC_TURNSTILE=disabled
 
 # PR previews
-NEXT_PUBLIC_PREVIEW_STEALING_DEV=
+NEXT_PUBLIC_PREVIEW_STEALING_DEV=

autogpt_platform/frontend/CONTRIBUTING.md

@@ -175,6 +175,8 @@ While server components and actions are cool and cutting-edge, they introduce a
 
 - Prefer [React Query](https://tanstack.com/query/latest/docs/framework/react/overview) for server state, colocated near consumers (see [state colocation](https://kentcdodds.com/blog/state-colocation-will-make-your-react-app-faster))
 - Co-locate UI state inside components/hooks; keep global state minimal
+- Avoid `useMemo` and `useCallback` unless you have a measured performance issue
+- Do not abuse `useEffect`; prefer state colocation and derive values directly when possible
 
 ### Styling and components
 
@@ -549,9 +551,48 @@ Files:
 Types:
 
 - Prefer `interface` for object shapes
-- Component props should be `interface Props { ... }`
+- Component props should be `interface Props { ... }` (not exported)
+- Only use specific exported names (e.g., `export interface MyComponentProps`) when the interface needs to be used outside the component
+- Keep type definitions inline with the component - do not create separate `types.ts` files unless types are shared across multiple files
 - Use precise types; avoid `any` and unsafe casts
 
+**Props naming examples:**
+
+```tsx
+// ✅ Good - internal props, not exported
+interface Props {
+  title: string;
+  onClose: () => void;
+}
+
+export function Modal({ title, onClose }: Props) {
+  // ...
+}
+
+// ✅ Good - exported when needed externally
+export interface ModalProps {
+  title: string;
+  onClose: () => void;
+}
+
+export function Modal({ title, onClose }: ModalProps) {
+  // ...
+}
+
+// ❌ Bad - unnecessarily specific name for internal use
+interface ModalComponentProps {
+  title: string;
+  onClose: () => void;
+}
+
+// ❌ Bad - separate types.ts file for single component
+// types.ts
+export interface ModalProps { ... }
+
+// Modal.tsx
+import type { ModalProps } from './types';
+```
+
 Parameters:
 
 - If more than one parameter is needed, pass a single `Args` object for clarity

autogpt_platform/frontend/orval.config.ts

@@ -16,6 +16,12 @@ export default defineConfig({
       client: "react-query",
       httpClient: "fetch",
       indexFiles: false,
+      mock: {
+        type: "msw",
+        baseUrl: "http://localhost:3000/api/proxy",
+        generateEachHttpStatus: true,
+        delay: 0,
+      },
       override: {
         mutator: {
           path: "./mutators/custom-mutator.ts",

autogpt_platform/frontend/package.json

@@ -15,6 +15,8 @@
     "types": "tsc --noEmit",
     "test": "NEXT_PUBLIC_PW_TEST=true next build --turbo && playwright test",
     "test-ui": "NEXT_PUBLIC_PW_TEST=true next build --turbo && playwright test --ui",
+    "test:unit": "vitest run",
+    "test:unit:watch": "vitest",
     "test:no-build": "playwright test",
     "gentests": "playwright codegen http://localhost:3000",
     "storybook": "storybook dev -p 6006",
@@ -118,6 +120,7 @@
   },
   "devDependencies": {
     "@chromatic-com/storybook": "4.1.2",
+    "happy-dom": "20.3.4",
     "@opentelemetry/instrumentation": "0.209.0",
     "@playwright/test": "1.56.1",
     "@storybook/addon-a11y": "9.1.5",
@@ -127,6 +130,8 @@
     "@storybook/nextjs": "9.1.5",
     "@tanstack/eslint-plugin-query": "5.91.2",
     "@tanstack/react-query-devtools": "5.90.2",
+    "@testing-library/dom": "10.4.1",
+    "@testing-library/react": "16.3.2",
     "@types/canvas-confetti": "1.9.0",
     "@types/lodash": "4.17.20",
     "@types/negotiator": "0.6.4",
@@ -135,6 +140,7 @@
     "@types/react-dom": "18.3.5",
     "@types/react-modal": "3.16.3",
     "@types/react-window": "1.8.8",
+    "@vitejs/plugin-react": "5.1.2",
     "axe-playwright": "2.2.2",
     "chromatic": "13.3.3",
     "concurrently": "9.2.1",
@@ -153,7 +159,9 @@
     "require-in-the-middle": "8.0.1",
     "storybook": "9.1.5",
     "tailwindcss": "3.4.17",
-    "typescript": "5.9.3"
+    "typescript": "5.9.3",
+    "vite-tsconfig-paths": "6.0.4",
+    "vitest": "4.0.17"
   },
   "msw": {
     "workerDirectory": [

autogpt_platform/frontend/src/app/(no-navbar)/logout/page.tsx

@@ -0,0 +1,58 @@
+"use client";
+
+import { LoadingSpinner } from "@/components/atoms/LoadingSpinner/LoadingSpinner";
+import { Text } from "@/components/atoms/Text/Text";
+import { useToast } from "@/components/molecules/Toast/use-toast";
+import { useSupabase } from "@/lib/supabase/hooks/useSupabase";
+import { useRouter } from "next/navigation";
+import { useEffect, useRef } from "react";
+
+const LOGOUT_REDIRECT_DELAY_MS = 400;
+
+function wait(ms: number): Promise<void> {
+  return new Promise(function resolveAfterDelay(resolve) {
+    setTimeout(resolve, ms);
+  });
+}
+
+export default function LogoutPage() {
+  const { logOut } = useSupabase();
+  const { toast } = useToast();
+  const router = useRouter();
+  const hasStartedRef = useRef(false);
+
+  useEffect(
+    function handleLogoutEffect() {
+      if (hasStartedRef.current) return;
+      hasStartedRef.current = true;
+
+      async function runLogout() {
+        try {
+          await logOut();
+        } catch {
+          toast({
+            title: "Failed to log out. Redirecting to login.",
+            variant: "destructive",
+          });
+        } finally {
+          await wait(LOGOUT_REDIRECT_DELAY_MS);
+          router.replace("/login");
+        }
+      }
+
+      void runLogout();
+    },
+    [logOut, router, toast],
+  );
+
+  return (
+    <div className="flex min-h-screen items-center justify-center px-4">
+      <div className="flex flex-col items-center justify-center gap-4 py-8">
+        <LoadingSpinner size="large" />
+        <Text variant="body" className="text-center">
+          Logging you out...
+        </Text>
+      </div>
+    </div>
+  );
+}

autogpt_platform/frontend/src/app/(platform)/auth/callback/route.ts

@@ -9,7 +9,7 @@ export async function GET(request: Request) {
   const { searchParams, origin } = new URL(request.url);
   const code = searchParams.get("code");
 
-  let next = "/marketplace";
+  let next = "/";
 
   if (code) {
     const supabase = await getServerSupabase();

autogpt_platform/frontend/src/app/(platform)/build/components/BuilderActions/components/RunGraph/RunGraph.tsx

@@ -5,10 +5,11 @@ import {
   TooltipContent,
   TooltipTrigger,
 } from "@/components/atoms/Tooltip/BaseTooltip";
-import { PlayIcon, StopIcon } from "@phosphor-icons/react";
+import { CircleNotchIcon, PlayIcon, StopIcon } from "@phosphor-icons/react";
 import { useShallow } from "zustand/react/shallow";
 import { RunInputDialog } from "../RunInputDialog/RunInputDialog";
 import { useRunGraph } from "./useRunGraph";
+import { cn } from "@/lib/utils";
 
 export const RunGraph = ({ flowID }: { flowID: string | null }) => {
   const {
@@ -24,6 +25,31 @@ export const RunGraph = ({ flowID }: { flowID: string | null }) => {
     useShallow((state) => state.isGraphRunning),
   );
 
+  const isLoading = isExecutingGraph || isTerminatingGraph || isSaving;
+
+  // Determine which icon to show with proper animation
+  const renderIcon = () => {
+    const iconClass = cn(
+      "size-4 transition-transform duration-200 ease-out",
+      !isLoading && "group-hover:scale-110",
+    );
+
+    if (isLoading) {
+      return (
+        <CircleNotchIcon
+          className={cn(iconClass, "animate-spin")}
+          weight="bold"
+        />
+      );
+    }
+
+    if (isGraphRunning) {
+      return <StopIcon className={iconClass} weight="fill" />;
+    }
+
+    return <PlayIcon className={iconClass} weight="fill" />;
+  };
+
   return (
     <>
       <Tooltip>
@@ -33,18 +59,18 @@ export const RunGraph = ({ flowID }: { flowID: string | null }) => {
             variant={isGraphRunning ? "destructive" : "primary"}
             data-id={isGraphRunning ? "stop-graph-button" : "run-graph-button"}
             onClick={isGraphRunning ? handleStopGraph : handleRunGraph}
-            disabled={!flowID || isExecutingGraph || isTerminatingGraph}
-            loading={isExecutingGraph || isTerminatingGraph || isSaving}
+            disabled={!flowID || isLoading}
+            className="group"
           >
-            {!isGraphRunning ? (
-              <PlayIcon className="size-4" />
-            ) : (
-              <StopIcon className="size-4" />
-            )}
+            {renderIcon()}
           </Button>
         </TooltipTrigger>
         <TooltipContent>
-          {isGraphRunning ? "Stop agent" : "Run agent"}
+          {isLoading
+            ? "Processing..."
+            : isGraphRunning
+              ? "Stop agent"
+              : "Run agent"}
         </TooltipContent>
       </Tooltip>
       <RunInputDialog

autogpt_platform/frontend/src/app/(platform)/build/components/BuilderActions/components/RunInputDialog/RunInputDialog.tsx

@@ -10,6 +10,7 @@ import { useRunInputDialog } from "./useRunInputDialog";
 import { CronSchedulerDialog } from "../CronSchedulerDialog/CronSchedulerDialog";
 import { useTutorialStore } from "@/app/(platform)/build/stores/tutorialStore";
 import { useEffect } from "react";
+import { CredentialsGroupedView } from "@/components/contextual/CredentialsInput/components/CredentialsGroupedView/CredentialsGroupedView";
 
 export const RunInputDialog = ({
   isOpen,
@@ -23,19 +24,17 @@ export const RunInputDialog = ({
   const hasInputs = useGraphStore((state) => state.hasInputs);
   const hasCredentials = useGraphStore((state) => state.hasCredentials);
   const inputSchema = useGraphStore((state) => state.inputSchema);
-  const credentialsSchema = useGraphStore(
-    (state) => state.credentialsInputSchema,
-  );
 
   const {
-    credentialsUiSchema,
+    credentialFields,
+    requiredCredentials,
     handleManualRun,
     handleInputChange,
     openCronSchedulerDialog,
     setOpenCronSchedulerDialog,
     inputValues,
     credentialValues,
-    handleCredentialChange,
+    handleCredentialFieldChange,
     isExecutingGraph,
   } = useRunInputDialog({ setIsOpen });
 
@@ -62,67 +61,67 @@ export const RunInputDialog = ({
           isOpen,
           set: setIsOpen,
         }}
-        styling={{ maxWidth: "600px", minWidth: "600px" }}
+        styling={{ maxWidth: "700px", minWidth: "700px" }}
       >
         <Dialog.Content>
-          <div className="space-y-6 p-1" data-id="run-input-dialog-content">
-            {/* Credentials Section */}
-            {hasCredentials() && (
-              <div data-id="run-input-credentials-section">
-                <div className="mb-4">
-                  <Text variant="h4" className="text-gray-900">
-                    Credentials
-                  </Text>
+          <div
+            className="grid grid-cols-[1fr_auto] gap-10 p-1"
+            data-id="run-input-dialog-content"
+          >
+            <div className="space-y-6">
+              {/* Credentials Section */}
+              {hasCredentials() && credentialFields.length > 0 && (
+                <div data-id="run-input-credentials-section">
+                  <div className="mb-4">
+                    <Text variant="h4" className="text-gray-900">
+                      Credentials
+                    </Text>
+                  </div>
+                  <div className="px-2" data-id="run-input-credentials-form">
+                    <CredentialsGroupedView
+                      credentialFields={credentialFields}
+                      requiredCredentials={requiredCredentials}
+                      inputCredentials={credentialValues}
+                      inputValues={inputValues}
+                      onCredentialChange={handleCredentialFieldChange}
+                    />
+                  </div>
                 </div>
-                <div className="px-2" data-id="run-input-credentials-form">
-                  <FormRenderer
-                    jsonSchema={credentialsSchema as RJSFSchema}
-                    handleChange={(v) => handleCredentialChange(v.formData)}
-                    uiSchema={credentialsUiSchema}
-                    initialValues={{}}
-                    formContext={{
-                      showHandles: false,
-                      size: "large",
-                      showOptionalToggle: false,
-                    }}
-                  />
-                </div>
-              </div>
-            )}
+              )}
 
-            {/* Inputs Section */}
-            {hasInputs() && (
-              <div data-id="run-input-inputs-section">
-                <div className="mb-4">
-                  <Text variant="h4" className="text-gray-900">
-                    Inputs
-                  </Text>
+              {/* Inputs Section */}
+              {hasInputs() && (
+                <div data-id="run-input-inputs-section">
+                  <div className="mb-4">
+                    <Text variant="h4" className="text-gray-900">
+                      Inputs
+                    </Text>
+                  </div>
+                  <div data-id="run-input-inputs-form">
+                    <FormRenderer
+                      jsonSchema={inputSchema as RJSFSchema}
+                      handleChange={(v) => handleInputChange(v.formData)}
+                      uiSchema={uiSchema}
+                      initialValues={{}}
+                      formContext={{
+                        showHandles: false,
+                        size: "large",
+                      }}
+                    />
+                  </div>
                 </div>
-                <div data-id="run-input-inputs-form">
-                  <FormRenderer
-                    jsonSchema={inputSchema as RJSFSchema}
-                    handleChange={(v) => handleInputChange(v.formData)}
-                    uiSchema={uiSchema}
-                    initialValues={{}}
-                    formContext={{
-                      showHandles: false,
-                      size: "large",
-                    }}
-                  />
-                </div>
-              </div>
-            )}
+              )}
+            </div>
 
-            {/* Action Button */}
             <div
-              className="flex justify-end pt-2"
+              className="flex flex-col items-end justify-start"
               data-id="run-input-actions-section"
             >
               {purpose === "run" && (
                 <Button
                   variant="primary"
                   size="large"
-                  className="group h-fit min-w-0 gap-2"
+                  className="group h-fit min-w-0 gap-2 px-10"
                   onClick={handleManualRun}
                   loading={isExecutingGraph}
                   data-id="run-input-manual-run-button"
@@ -137,7 +136,7 @@ export const RunInputDialog = ({
                 <Button
                   variant="primary"
                   size="large"
-                  className="group h-fit min-w-0 gap-2"
+                  className="group h-fit min-w-0 gap-2 px-10"
                   onClick={() => setOpenCronSchedulerDialog(true)}
                   data-id="run-input-schedule-button"
                 >

autogpt_platform/frontend/src/app/(platform)/build/components/BuilderActions/components/RunInputDialog/useRunInputDialog.ts

@@ -7,12 +7,11 @@ import {
   GraphExecutionMeta,
 } from "@/lib/autogpt-server-api";
 import { parseAsInteger, parseAsString, useQueryStates } from "nuqs";
-import { useMemo, useState } from "react";
-import { uiSchema } from "../../../FlowEditor/nodes/uiSchema";
-import { isCredentialFieldSchema } from "@/components/renderers/InputRenderer/custom/CredentialField/helpers";
+import { useCallback, useMemo, useState } from "react";
 import { useNodeStore } from "@/app/(platform)/build/stores/nodeStore";
 import { useToast } from "@/components/molecules/Toast/use-toast";
 import { useReactFlow } from "@xyflow/react";
+import type { CredentialField } from "@/components/contextual/CredentialsInput/components/CredentialsGroupedView/helpers";
 
 export const useRunInputDialog = ({
   setIsOpen,
@@ -120,27 +119,32 @@ export const useRunInputDialog = ({
       },
     });
 
-  // We are rendering the credentials field differently compared to other fields.
-  // In the node, we have the field name as "credential" - so our library catches it and renders it differently.
-  // But here we have a different name, something like `Firecrawl credentials`, so here we are telling the library that this field is a credential field type.
+  // Convert credentials schema to credential fields array for CredentialsGroupedView
+  const credentialFields: CredentialField[] = useMemo(() => {
+    if (!credentialsSchema?.properties) return [];
+    return Object.entries(credentialsSchema.properties);
+  }, [credentialsSchema]);
 
-  const credentialsUiSchema = useMemo(() => {
-    const dynamicUiSchema: any = { ...uiSchema };
+  // Get required credentials as a Set
+  const requiredCredentials = useMemo(() => {
+    return new Set<string>(credentialsSchema?.required || []);
+  }, [credentialsSchema]);
 
-    if (credentialsSchema?.properties) {
-      Object.keys(credentialsSchema.properties).forEach((fieldName) => {
-        const fieldSchema = credentialsSchema.properties[fieldName];
-        if (isCredentialFieldSchema(fieldSchema)) {
-          dynamicUiSchema[fieldName] = {
-            ...dynamicUiSchema[fieldName],
-            "ui:field": "custom/credential_field",
-          };
+  // Handler for individual credential changes
+  const handleCredentialFieldChange = useCallback(
+    (key: string, value?: CredentialsMetaInput) => {
+      setCredentialValues((prev) => {
+        if (value) {
+          return { ...prev, [key]: value };
+        } else {
+          const next = { ...prev };
+          delete next[key];
+          return next;
         }
       });
-    }
-
-    return dynamicUiSchema;
-  }, [credentialsSchema]);
+    },
+    [],
+  );
 
   const handleManualRun = async () => {
     // Filter out incomplete credentials (those without a valid id)
@@ -173,12 +177,14 @@ export const useRunInputDialog = ({
   };
 
   return {
-    credentialsUiSchema,
+    credentialFields,
+    requiredCredentials,
     inputValues,
     credentialValues,
     isExecutingGraph,
     handleInputChange,
     handleCredentialChange,
+    handleCredentialFieldChange,
     handleManualRun,
     openCronSchedulerDialog,
     setOpenCronSchedulerDialog,

autogpt_platform/frontend/src/app/(platform)/build/components/FloatingSafeModeToogle.tsx

@@ -18,69 +18,118 @@ interface Props {
   fullWidth?: boolean;
 }
 
+interface SafeModeButtonProps {
+  isEnabled: boolean;
+  label: string;
+  tooltipEnabled: string;
+  tooltipDisabled: string;
+  onToggle: () => void;
+  isPending: boolean;
+  fullWidth?: boolean;
+}
+
+function SafeModeButton({
+  isEnabled,
+  label,
+  tooltipEnabled,
+  tooltipDisabled,
+  onToggle,
+  isPending,
+  fullWidth = false,
+}: SafeModeButtonProps) {
+  return (
+    <Tooltip delayDuration={100}>
+      <TooltipTrigger asChild>
+        <Button
+          variant={isEnabled ? "primary" : "outline"}
+          size="small"
+          onClick={onToggle}
+          disabled={isPending}
+          className={cn("justify-start", fullWidth ? "w-full" : "")}
+        >
+          {isEnabled ? (
+            <>
+              <ShieldCheckIcon weight="bold" size={16} />
+              <Text variant="body" className="text-zinc-200">
+                {label}: ON
+              </Text>
+            </>
+          ) : (
+            <>
+              <ShieldIcon weight="bold" size={16} />
+              <Text variant="body" className="text-zinc-600">
+                {label}: OFF
+              </Text>
+            </>
+          )}
+        </Button>
+      </TooltipTrigger>
+      <TooltipContent>
+        <div className="text-center">
+          <div className="font-medium">
+            {label}: {isEnabled ? "ON" : "OFF"}
+          </div>
+          <div className="mt-1 text-xs text-muted-foreground">
+            {isEnabled ? tooltipEnabled : tooltipDisabled}
+          </div>
+        </div>
+      </TooltipContent>
+    </Tooltip>
+  );
+}
+
 export function FloatingSafeModeToggle({
   graph,
   className,
   fullWidth = false,
 }: Props) {
   const {
-    currentSafeMode,
+    currentHITLSafeMode,
+    showHITLToggle,
+    isHITLStateUndetermined,
+    handleHITLToggle,
+    currentSensitiveActionSafeMode,
+    showSensitiveActionToggle,
+    handleSensitiveActionToggle,
     isPending,
     shouldShowToggle,
-    isStateUndetermined,
-    handleToggle,
   } = useAgentSafeMode(graph);
 
-  if (!shouldShowToggle || isStateUndetermined || isPending) {
+  if (!shouldShowToggle || isPending) {
+    return null;
+  }
+
+  const showHITL = showHITLToggle && !isHITLStateUndetermined;
+  const showSensitive = showSensitiveActionToggle;
+
+  if (!showHITL && !showSensitive) {
     return null;
   }
 
   return (
-    <div className={cn("fixed z-50", className)}>
-      <Tooltip delayDuration={100}>
-        <TooltipTrigger asChild>
-          <Button
-            variant={currentSafeMode! ? "primary" : "outline"}
-            key={graph.id}
-            size="small"
-            title={
-              currentSafeMode!
-                ? "Safe Mode: ON. Human in the loop blocks require manual review"
-                : "Safe Mode: OFF. Human in the loop blocks proceed automatically"
-            }
-            onClick={handleToggle}
-            className={cn(fullWidth ? "w-full" : "")}
-          >
-            {currentSafeMode! ? (
-              <>
-                <ShieldCheckIcon weight="bold" size={16} />
-                <Text variant="body" className="text-zinc-200">
-                  Safe Mode: ON
-                </Text>
-              </>
-            ) : (
-              <>
-                <ShieldIcon weight="bold" size={16} />
-                <Text variant="body" className="text-zinc-600">
-                  Safe Mode: OFF
-                </Text>
-              </>
-            )}
-          </Button>
-        </TooltipTrigger>
-        <TooltipContent>
-          <div className="text-center">
-            <div className="font-medium">
-              Safe Mode: {currentSafeMode! ? "ON" : "OFF"}
-            </div>
-            <div className="mt-1 text-xs text-muted-foreground">
-              {currentSafeMode!
-                ? "Human in the loop blocks require manual review"
-                : "Human in the loop blocks proceed automatically"}
-            </div>
-          </div>
-        </TooltipContent>
-      </Tooltip>
+    <div className={cn("fixed z-50 flex flex-col gap-2", className)}>
+      {showHITL && (
+        <SafeModeButton
+          isEnabled={currentHITLSafeMode}
+          label="Human in the loop block approval"
+          tooltipEnabled="The agent will pause at human-in-the-loop blocks and wait for your approval"
+          tooltipDisabled="Human in the loop blocks will proceed automatically"
+          onToggle={handleHITLToggle}
+          isPending={isPending}
+          fullWidth={fullWidth}
+        />
+      )}
+      {showSensitive && (
+        <SafeModeButton
+          isEnabled={currentSensitiveActionSafeMode}
+          label="Sensitive actions blocks approval"
+          tooltipEnabled="The agent will pause at sensitive action blocks and wait for your approval"
+          tooltipDisabled="Sensitive action blocks will proceed automatically"
+          onToggle={handleSensitiveActionToggle}
+          isPending={isPending}
+          fullWidth={fullWidth}
+        />
+      )}
     </div>
   );
 }

autogpt_platform/frontend/src/app/(platform)/build/components/FlowEditor/Flow/components/CustomControl.tsx

@@ -53,14 +53,14 @@ export const CustomControls = memo(
     const controls = [
       {
         id: "zoom-in-button",
-        icon: <PlusIcon className="size-4" />,
+        icon: <PlusIcon className="size-3.5 text-zinc-600" />,
         label: "Zoom In",
         onClick: () => zoomIn(),
         className: "h-10 w-10 border-none",
       },
       {
         id: "zoom-out-button",
-        icon: <MinusIcon className="size-4" />,
+        icon: <MinusIcon className="size-3.5 text-zinc-600" />,
         label: "Zoom Out",
         onClick: () => zoomOut(),
         className: "h-10 w-10 border-none",
@@ -68,9 +68,9 @@ export const CustomControls = memo(
       {
         id: "tutorial-button",
         icon: isTutorialLoading ? (
-          <CircleNotchIcon className="size-4 animate-spin" />
+          <CircleNotchIcon className="size-3.5 animate-spin text-zinc-600" />
         ) : (
-          <ChalkboardIcon className="size-4" />
+          <ChalkboardIcon className="size-3.5 text-zinc-600" />
         ),
         label: isTutorialLoading ? "Loading Tutorial..." : "Start Tutorial",
         onClick: handleTutorialClick,
@@ -79,7 +79,7 @@ export const CustomControls = memo(
       },
       {
         id: "fit-view-button",
-        icon: <FrameCornersIcon className="size-4" />,
+        icon: <FrameCornersIcon className="size-3.5 text-zinc-600" />,
         label: "Fit View",
         onClick: () => fitView({ padding: 0.2, duration: 800, maxZoom: 1 }),
         className: "h-10 w-10 border-none",
@@ -87,9 +87,9 @@ export const CustomControls = memo(
       {
         id: "lock-button",
         icon: !isLocked ? (
-          <LockOpenIcon className="size-4" />
+          <LockOpenIcon className="size-3.5 text-zinc-600" />
         ) : (
-          <LockIcon className="size-4" />
+          <LockIcon className="size-3.5 text-zinc-600" />
         ),
         label: "Toggle Lock",
         onClick: () => setIsLocked(!isLocked),

autogpt_platform/frontend/src/app/(platform)/build/components/FlowEditor/Flow/useFlow.ts

@@ -139,14 +139,6 @@ export const useFlow = () => {
       useNodeStore.getState().setNodes([]);
       useNodeStore.getState().clearResolutionState();
       addNodes(customNodes);
-
-      // Sync hardcoded values with handle IDs.
-      // If a key–value field has a key without a value, the backend omits it from hardcoded values.
-      // But if a handleId exists for that key, it causes inconsistency.
-      // This ensures hardcoded values stay in sync with handle IDs.
-      customNodes.forEach((node) => {
-        useNodeStore.getState().syncHardcodedValuesWithHandleIds(node.id);
-      });
     }
   }, [customNodes, addNodes]);
 
@@ -158,6 +150,14 @@ export const useFlow = () => {
     }
   }, [graph?.links, addLinks]);
 
+  useEffect(() => {
+    if (customNodes.length > 0 && graph?.links) {
+      customNodes.forEach((node) => {
+        useNodeStore.getState().syncHardcodedValuesWithHandleIds(node.id);
+      });
+    }
+  }, [customNodes, graph?.links]);
+
   // update node execution status in nodes
   useEffect(() => {
     if (

autogpt_platform/frontend/src/app/(platform)/build/components/FlowEditor/edges/CustomEdge.tsx

@@ -19,6 +19,8 @@ export type CustomEdgeData = {
   beadUp?: number;
   beadDown?: number;
   beadData?: Map<string, NodeExecutionResult["status"]>;
+  edgeColorClass?: string;
+  edgeHexColor?: string;
 };
 
 export type CustomEdge = XYEdge<CustomEdgeData, "custom">;
@@ -36,7 +38,6 @@ const CustomEdge = ({
   selected,
 }: EdgeProps<CustomEdge>) => {
   const removeConnection = useEdgeStore((state) => state.removeEdge);
-  // Subscribe to the brokenEdgeIDs map and check if this edge is broken across any node
   const isBroken = useNodeStore((state) => state.isEdgeBroken(id));
   const [isHovered, setIsHovered] = useState(false);
 
@@ -52,6 +53,7 @@ const CustomEdge = ({
   const isStatic = data?.isStatic ?? false;
   const beadUp = data?.beadUp ?? 0;
   const beadDown = data?.beadDown ?? 0;
+  const edgeColorClass = data?.edgeColorClass;
 
   const handleRemoveEdge = () => {
     removeConnection(id);
@@ -70,7 +72,9 @@ const CustomEdge = ({
             ? "!stroke-red-500 !stroke-[2px] [stroke-dasharray:4]"
             : selected
               ? "stroke-zinc-800"
-              : "stroke-zinc-500/50 hover:stroke-zinc-500",
+              : edgeColorClass
+                ? cn(edgeColorClass, "opacity-70 hover:opacity-100")
+                : "stroke-zinc-500/50 hover:stroke-zinc-500",
         )}
       />
       <JSBeads

autogpt_platform/frontend/src/app/(platform)/build/components/FlowEditor/edges/useCustomEdge.ts

@@ -8,6 +8,7 @@ import { useCallback } from "react";
 import { useNodeStore } from "../../../stores/nodeStore";
 import { useHistoryStore } from "../../../stores/historyStore";
 import { CustomEdge } from "./CustomEdge";
+import { getEdgeColorFromOutputType } from "../nodes/helpers";
 
 export const useCustomEdge = () => {
   const edges = useEdgeStore((s) => s.edges);
@@ -34,8 +35,13 @@ export const useCustomEdge = () => {
       if (exists) return;
 
       const nodes = useNodeStore.getState().nodes;
-      const isStatic = nodes.find((n) => n.id === conn.source)?.data
-        ?.staticOutput;
+      const sourceNode = nodes.find((n) => n.id === conn.source);
+      const isStatic = sourceNode?.data?.staticOutput;
+
+      const { colorClass, hexColor } = getEdgeColorFromOutputType(
+        sourceNode?.data?.outputSchema,
+        conn.sourceHandle,
+      );
 
       addEdge({
         source: conn.source,
@@ -44,6 +50,8 @@ export const useCustomEdge = () => {
         targetHandle: conn.targetHandle,
         data: {
           isStatic,
+          edgeColorClass: colorClass,
+          edgeHexColor: hexColor,
         },
       });
     },

autogpt_platform/frontend/src/app/(platform)/build/components/FlowEditor/nodes/CustomNode/components/NodeOutput/NodeOutput.tsx

@@ -1,22 +1,21 @@
 import { Button } from "@/components/atoms/Button/Button";
 import { Text } from "@/components/atoms/Text/Text";
+import {
+  Accordion,
+  AccordionContent,
+  AccordionItem,
+  AccordionTrigger,
+} from "@/components/molecules/Accordion/Accordion";
 import { beautifyString, cn } from "@/lib/utils";
-import { CaretDownIcon, CopyIcon, CheckIcon } from "@phosphor-icons/react";
+import { CopyIcon, CheckIcon } from "@phosphor-icons/react";
 import { NodeDataViewer } from "./components/NodeDataViewer/NodeDataViewer";
 import { ContentRenderer } from "./components/ContentRenderer";
 import { useNodeOutput } from "./useNodeOutput";
 import { ViewMoreData } from "./components/ViewMoreData";
 
 export const NodeDataRenderer = ({ nodeId }: { nodeId: string }) => {
-  const {
-    outputData,
-    isExpanded,
-    setIsExpanded,
-    copiedKey,
-    handleCopy,
-    executionResultId,
-    inputData,
-  } = useNodeOutput(nodeId);
+  const { outputData, copiedKey, handleCopy, executionResultId, inputData } =
+    useNodeOutput(nodeId);
 
   if (Object.keys(outputData).length === 0) {
     return null;
@@ -25,122 +24,117 @@ export const NodeDataRenderer = ({ nodeId }: { nodeId: string }) => {
   return (
     <div
       data-tutorial-id={`node-output`}
-      className="flex flex-col gap-3 rounded-b-xl border-t border-zinc-200 px-4 py-4"
+      className="rounded-b-xl border-t border-zinc-200 px-4 py-2"
     >
-      <div className="flex items-center justify-between">
-        <Text variant="body-medium" className="!font-semibold text-slate-700">
-          Node Output
-        </Text>
-        <Button
-          variant="ghost"
-          size="small"
-          onClick={() => setIsExpanded(!isExpanded)}
-          className="h-fit min-w-0 p-1 text-slate-600 hover:text-slate-900"
-        >
-          <CaretDownIcon
-            size={16}
-            weight="bold"
-            className={`transition-transform ${isExpanded ? "rotate-180" : ""}`}
-          />
-        </Button>
-      </div>
+      <Accordion type="single" collapsible defaultValue="node-output">
+        <AccordionItem value="node-output" className="border-none">
+          <AccordionTrigger className="py-2 hover:no-underline">
+            <Text
+              variant="body-medium"
+              className="!font-semibold text-slate-700"
+            >
+              Node Output
+            </Text>
+          </AccordionTrigger>
+          <AccordionContent className="pt-2">
+            <div className="flex max-w-[350px] flex-col gap-4">
+              <div className="space-y-2">
+                <Text variant="small-medium">Input</Text>
 
-      {isExpanded && (
-        <>
-          <div className="flex max-w-[350px] flex-col gap-4">
-            <div className="space-y-2">
-              <Text variant="small-medium">Input</Text>
+                <ContentRenderer value={inputData} shortContent={false} />
 
-              <ContentRenderer value={inputData} shortContent={false} />
-
-              <div className="mt-1 flex justify-end gap-1">
-                <NodeDataViewer
-                  data={inputData}
-                  pinName="Input"
-                  execId={executionResultId}
-                />
-                <Button
-                  variant="secondary"
-                  size="small"
-                  onClick={() => handleCopy("input", inputData)}
-                  className={cn(
-                    "h-fit min-w-0 gap-1.5 border border-zinc-200 p-2 text-black hover:text-slate-900",
-                    copiedKey === "input" &&
-                      "border-green-400 bg-green-100 hover:border-green-400 hover:bg-green-200",
-                  )}
-                >
-                  {copiedKey === "input" ? (
-                    <CheckIcon size={12} className="text-green-600" />
-                  ) : (
-                    <CopyIcon size={12} />
-                  )}
-                </Button>
+                <div className="mt-1 flex justify-end gap-1">
+                  <NodeDataViewer
+                    data={inputData}
+                    pinName="Input"
+                    execId={executionResultId}
+                  />
+                  <Button
+                    variant="secondary"
+                    size="small"
+                    onClick={() => handleCopy("input", inputData)}
+                    className={cn(
+                      "h-fit min-w-0 gap-1.5 border border-zinc-200 p-2 text-black hover:text-slate-900",
+                      copiedKey === "input" &&
+                        "border-green-400 bg-green-100 hover:border-green-400 hover:bg-green-200",
+                    )}
+                  >
+                    {copiedKey === "input" ? (
+                      <CheckIcon size={12} className="text-green-600" />
+                    ) : (
+                      <CopyIcon size={12} />
+                    )}
+                  </Button>
+                </div>
               </div>
-            </div>
 
-            {Object.entries(outputData)
-              .slice(0, 2)
-              .map(([key, value]) => (
-                <div key={key} className="flex flex-col gap-2">
-                  <div className="flex items-center gap-2">
-                    <Text
-                      variant="small-medium"
-                      className="!font-semibold text-slate-600"
-                    >
-                      Pin:
-                    </Text>
-                    <Text variant="small" className="text-slate-700">
-                      {beautifyString(key)}
-                    </Text>
-                  </div>
-                  <div className="w-full space-y-2">
-                    <Text
-                      variant="small"
-                      className="!font-semibold text-slate-600"
-                    >
-                      Data:
-                    </Text>
-                    <div className="relative space-y-2">
-                      {value.map((item, index) => (
-                        <div key={index}>
-                          <ContentRenderer value={item} shortContent={true} />
+              {Object.entries(outputData)
+                .slice(0, 2)
+                .map(([key, value]) => (
+                  <div key={key} className="flex flex-col gap-2">
+                    <div className="flex items-center gap-2">
+                      <Text
+                        variant="small-medium"
+                        className="!font-semibold text-slate-600"
+                      >
+                        Pin:
+                      </Text>
+                      <Text variant="small" className="text-slate-700">
+                        {beautifyString(key)}
+                      </Text>
+                    </div>
+                    <div className="w-full space-y-2">
+                      <Text
+                        variant="small"
+                        className="!font-semibold text-slate-600"
+                      >
+                        Data:
+                      </Text>
+                      <div className="relative space-y-2">
+                        {value.map((item, index) => (
+                          <div key={index}>
+                            <ContentRenderer value={item} shortContent={true} />
+                          </div>
+                        ))}
+
+                        <div className="mt-1 flex justify-end gap-1">
+                          <NodeDataViewer
+                            data={value}
+                            pinName={key}
+                            execId={executionResultId}
+                          />
+                          <Button
+                            variant="secondary"
+                            size="small"
+                            onClick={() => handleCopy(key, value)}
+                            className={cn(
+                              "h-fit min-w-0 gap-1.5 border border-zinc-200 p-2 text-black hover:text-slate-900",
+                              copiedKey === key &&
+                                "border-green-400 bg-green-100 hover:border-green-400 hover:bg-green-200",
+                            )}
+                          >
+                            {copiedKey === key ? (
+                              <CheckIcon size={12} className="text-green-600" />
+                            ) : (
+                              <CopyIcon size={12} />
+                            )}
+                          </Button>
                         </div>
-                      ))}
-
-                      <div className="mt-1 flex justify-end gap-1">
-                        <NodeDataViewer
-                          data={value}
-                          pinName={key}
-                          execId={executionResultId}
-                        />
-                        <Button
-                          variant="secondary"
-                          size="small"
-                          onClick={() => handleCopy(key, value)}
-                          className={cn(
-                            "h-fit min-w-0 gap-1.5 border border-zinc-200 p-2 text-black hover:text-slate-900",
-                            copiedKey === key &&
-                              "border-green-400 bg-green-100 hover:border-green-400 hover:bg-green-200",
-                          )}
-                        >
-                          {copiedKey === key ? (
-                            <CheckIcon size={12} className="text-green-600" />
-                          ) : (
-                            <CopyIcon size={12} />
-                          )}
-                        </Button>
                       </div>
                     </div>
                   </div>
-                </div>
-              ))}
-          </div>
+                ))}
+            </div>
 
-          {Object.keys(outputData).length > 2 && (
-            <ViewMoreData outputData={outputData} execId={executionResultId} />
-          )}
-        </>
-      )}
+            {Object.keys(outputData).length > 2 && (
+              <ViewMoreData
+                outputData={outputData}
+                execId={executionResultId}
+              />
+            )}
+          </AccordionContent>
+        </AccordionItem>
+      </Accordion>
     </div>
   );
 };

autogpt_platform/frontend/src/app/(platform)/build/components/FlowEditor/nodes/CustomNode/components/NodeOutput/useNodeOutput.tsx

@@ -4,7 +4,6 @@ import { useShallow } from "zustand/react/shallow";
 import { useState } from "react";
 
 export const useNodeOutput = (nodeId: string) => {
-  const [isExpanded, setIsExpanded] = useState(true);
   const [copiedKey, setCopiedKey] = useState<string | null>(null);
   const { toast } = useToast();
 
@@ -37,13 +36,10 @@ export const useNodeOutput = (nodeId: string) => {
     }
   };
   return {
-    outputData: outputData,
-    inputData: inputData,
-    isExpanded: isExpanded,
-    setIsExpanded: setIsExpanded,
-    copiedKey: copiedKey,
-    setCopiedKey: setCopiedKey,
-    handleCopy: handleCopy,
+    outputData,
+    inputData,
+    copiedKey,
+    handleCopy,
     executionResultId: nodeExecutionResult?.node_exec_id,
   };
 };

autogpt_platform/frontend/src/app/(platform)/build/components/FlowEditor/nodes/helpers.ts

@@ -187,3 +187,38 @@ export const getTypeDisplayInfo = (schema: any) => {
     hexColor,
   };
 };
+
+export function getEdgeColorFromOutputType(
+  outputSchema: RJSFSchema | undefined,
+  sourceHandle: string,
+): { colorClass: string; hexColor: string } {
+  const defaultColor = {
+    colorClass: "stroke-zinc-500/50",
+    hexColor: "#6b7280",
+  };
+
+  if (!outputSchema?.properties) return defaultColor;
+
+  const properties = outputSchema.properties as Record<string, unknown>;
+  const handleParts = sourceHandle.split("_#_");
+  let currentSchema: Record<string, unknown> = properties;
+
+  for (let i = 0; i < handleParts.length; i++) {
+    const part = handleParts[i];
+    const fieldSchema = currentSchema[part] as Record<string, unknown>;
+    if (!fieldSchema) return defaultColor;
+
+    if (i === handleParts.length - 1) {
+      const { hexColor, colorClass } = getTypeDisplayInfo(fieldSchema);
+      return { colorClass: colorClass.replace("!text-", "stroke-"), hexColor };
+    }
+
+    if (fieldSchema.properties) {
+      currentSchema = fieldSchema.properties as Record<string, unknown>;
+    } else {
+      return defaultColor;
+    }
+  }
+
+  return defaultColor;
+}

autogpt_platform/frontend/src/app/(platform)/build/components/FlowEditor/tutorial/icons.ts

@@ -1,7 +1,32 @@
-// These are SVG Phosphor icons
+type IconOptions = {
+  size?: number;
+  color?: string;
+};
+
+const DEFAULT_SIZE = 16;
+const DEFAULT_COLOR = "#52525b"; // zinc-600
+
+const iconPaths = {
+  ClickIcon: `M88,24V16a8,8,0,0,1,16,0v8a8,8,0,0,1-16,0ZM16,104h8a8,8,0,0,0,0-16H16a8,8,0,0,0,0,16ZM124.42,39.16a8,8,0,0,0,10.74-3.58l8-16a8,8,0,0,0-14.31-7.16l-8,16A8,8,0,0,0,124.42,39.16Zm-96,81.69-16,8a8,8,0,0,0,7.16,14.31l16-8a8,8,0,1,0-7.16-14.31ZM219.31,184a16,16,0,0,1,0,22.63l-12.68,12.68a16,16,0,0,1-22.63,0L132.7,168,115,214.09c0,.1-.08.21-.13.32a15.83,15.83,0,0,1-14.6,9.59l-.79,0a15.83,15.83,0,0,1-14.41-11L32.8,52.92A16,16,0,0,1,52.92,32.8L213,85.07a16,16,0,0,1,1.41,29.8l-.32.13L168,132.69ZM208,195.31,156.69,144h0a16,16,0,0,1,4.93-26l.32-.14,45.95-17.64L48,48l52.2,159.86,17.65-46c0-.11.08-.22.13-.33a16,16,0,0,1,11.69-9.34,16.72,16.72,0,0,1,3-.28,16,16,0,0,1,11.3,4.69L195.31,208Z`,
+  Keyboard: `M224,48H32A16,16,0,0,0,16,64V192a16,16,0,0,0,16,16H224a16,16,0,0,0,16-16V64A16,16,0,0,0,224,48Zm0,144H32V64H224V192Zm-16-64a8,8,0,0,1-8,8H56a8,8,0,0,1,0-16H200A8,8,0,0,1,208,128Zm0-32a8,8,0,0,1-8,8H56a8,8,0,0,1,0-16H200A8,8,0,0,1,208,96ZM72,160a8,8,0,0,1-8,8H56a8,8,0,0,1,0-16h8A8,8,0,0,1,72,160Zm96,0a8,8,0,0,1-8,8H96a8,8,0,0,1,0-16h64A8,8,0,0,1,168,160Zm40,0a8,8,0,0,1-8,8h-8a8,8,0,0,1,0-16h8A8,8,0,0,1,208,160Z`,
+  Drag: `M188,80a27.79,27.79,0,0,0-13.36,3.4,28,28,0,0,0-46.64-11A28,28,0,0,0,80,92v20H68a28,28,0,0,0-28,28v12a88,88,0,0,0,176,0V108A28,28,0,0,0,188,80Zm12,72a72,72,0,0,1-144,0V140a12,12,0,0,1,12-12H80v24a8,8,0,0,0,16,0V92a12,12,0,0,1,24,0v28a8,8,0,0,0,16,0V92a12,12,0,0,1,24,0v28a8,8,0,0,0,16,0V108a12,12,0,0,1,24,0Z`,
+};
+
+function createIcon(path: string, options: IconOptions = {}): string {
+  const size = options.size ?? DEFAULT_SIZE;
+  const color = options.color ?? DEFAULT_COLOR;
+  return `<svg xmlns="http://www.w3.org/2000/svg" width="${size}" height="${size}" fill="${color}" viewBox="0 0 256 256"><path d="${path}"></path></svg>`;
+}
 
 export const ICONS = {
-  ClickIcon: `<svg xmlns="http://www.w3.org/2000/svg" width="20" height="20" fill="#000000" viewBox="0 0 256 256"><path d="M88,24V16a8,8,0,0,1,16,0v8a8,8,0,0,1-16,0ZM16,104h8a8,8,0,0,0,0-16H16a8,8,0,0,0,0,16ZM124.42,39.16a8,8,0,0,0,10.74-3.58l8-16a8,8,0,0,0-14.31-7.16l-8,16A8,8,0,0,0,124.42,39.16Zm-96,81.69-16,8a8,8,0,0,0,7.16,14.31l16-8a8,8,0,1,0-7.16-14.31ZM219.31,184a16,16,0,0,1,0,22.63l-12.68,12.68a16,16,0,0,1-22.63,0L132.7,168,115,214.09c0,.1-.08.21-.13.32a15.83,15.83,0,0,1-14.6,9.59l-.79,0a15.83,15.83,0,0,1-14.41-11L32.8,52.92A16,16,0,0,1,52.92,32.8L213,85.07a16,16,0,0,1,1.41,29.8l-.32.13L168,132.69ZM208,195.31,156.69,144h0a16,16,0,0,1,4.93-26l.32-.14,45.95-17.64L48,48l52.2,159.86,17.65-46c0-.11.08-.22.13-.33a16,16,0,0,1,11.69-9.34,16.72,16.72,0,0,1,3-.28,16,16,0,0,1,11.3,4.69L195.31,208Z"></path></svg>`,
-  Keyboard: `<svg xmlns="http://www.w3.org/2000/svg" width="20" height="20" fill="#000000" viewBox="0 0 256 256"><path d="M224,48H32A16,16,0,0,0,16,64V192a16,16,0,0,0,16,16H224a16,16,0,0,0,16-16V64A16,16,0,0,0,224,48Zm0,144H32V64H224V192Zm-16-64a8,8,0,0,1-8,8H56a8,8,0,0,1,0-16H200A8,8,0,0,1,208,128Zm0-32a8,8,0,0,1-8,8H56a8,8,0,0,1,0-16H200A8,8,0,0,1,208,96ZM72,160a8,8,0,0,1-8,8H56a8,8,0,0,1,0-16h8A8,8,0,0,1,72,160Zm96,0a8,8,0,0,1-8,8H96a8,8,0,0,1,0-16h64A8,8,0,0,1,168,160Zm40,0a8,8,0,0,1-8,8h-8a8,8,0,0,1,0-16h8A8,8,0,0,1,208,160Z"></path></svg>`,
-  Drag: `<svg xmlns="http://www.w3.org/2000/svg" width="20" height="20" fill="#000000" viewBox="0 0 256 256"><path d="M188,80a27.79,27.79,0,0,0-13.36,3.4,28,28,0,0,0-46.64-11A28,28,0,0,0,80,92v20H68a28,28,0,0,0-28,28v12a88,88,0,0,0,176,0V108A28,28,0,0,0,188,80Zm12,72a72,72,0,0,1-144,0V140a12,12,0,0,1,12-12H80v24a8,8,0,0,0,16,0V92a12,12,0,0,1,24,0v28a8,8,0,0,0,16,0V92a12,12,0,0,1,24,0v28a8,8,0,0,0,16,0V108a12,12,0,0,1,24,0Z"></path></svg>`,
+  ClickIcon: createIcon(iconPaths.ClickIcon),
+  Keyboard: createIcon(iconPaths.Keyboard),
+  Drag: createIcon(iconPaths.Drag),
 };
+
+export function getIcon(
+  name: keyof typeof iconPaths,
+  options?: IconOptions,
+): string {
+  return createIcon(iconPaths[name], options);
+}

autogpt_platform/frontend/src/app/(platform)/build/components/FlowEditor/tutorial/index.ts

@@ -11,6 +11,7 @@ import {
 } from "./helpers";
 import { useNodeStore } from "../../../stores/nodeStore";
 import { useEdgeStore } from "../../../stores/edgeStore";
+import { useTutorialStore } from "../../../stores/tutorialStore";
 
 let isTutorialLoading = false;
 let tutorialLoadingCallback: ((loading: boolean) => void) | null = null;
@@ -60,12 +61,14 @@ export const startTutorial = async () => {
     handleTutorialComplete();
     removeTutorialStyles();
     clearPrefetchedBlocks();
+    useTutorialStore.getState().setIsTutorialRunning(false);
   });
 
   tour.on("cancel", () => {
     handleTutorialCancel(tour);
     removeTutorialStyles();
     clearPrefetchedBlocks();
+    useTutorialStore.getState().setIsTutorialRunning(false);
   });
 
   for (const step of tour.steps) {

autogpt_platform/frontend/src/app/(platform)/build/components/helper.ts

@@ -61,12 +61,18 @@ export const convertNodesPlusBlockInfoIntoCustomNodes = (
   return customNode;
 };
 
+const isToolSourceName = (sourceName: string): boolean =>
+  sourceName.startsWith("tools_^_");
+
+const cleanupSourceName = (sourceName: string): string =>
+  isToolSourceName(sourceName) ? "tools" : sourceName;
+
 export const linkToCustomEdge = (link: Link): CustomEdge => ({
   id: link.id ?? "",
   type: "custom" as const,
   source: link.source_id,
   target: link.sink_id,
-  sourceHandle: link.source_name,
+  sourceHandle: cleanupSourceName(link.source_name),
   targetHandle: link.sink_name,
   data: {
     isStatic: link.is_static,

autogpt_platform/frontend/src/app/(platform)/chat/components/Chat/Chat.tsx

@@ -1,134 +0,0 @@
-"use client";
-
-import { Button } from "@/components/atoms/Button/Button";
-import { Text } from "@/components/atoms/Text/Text";
-import { cn } from "@/lib/utils";
-import { List } from "@phosphor-icons/react";
-import React, { useState } from "react";
-import { ChatContainer } from "./components/ChatContainer/ChatContainer";
-import { ChatErrorState } from "./components/ChatErrorState/ChatErrorState";
-import { ChatLoadingState } from "./components/ChatLoadingState/ChatLoadingState";
-import { SessionsDrawer } from "./components/SessionsDrawer/SessionsDrawer";
-import { useChat } from "./useChat";
-
-export interface ChatProps {
-  className?: string;
-  headerTitle?: React.ReactNode;
-  showHeader?: boolean;
-  showSessionInfo?: boolean;
-  showNewChatButton?: boolean;
-  onNewChat?: () => void;
-  headerActions?: React.ReactNode;
-}
-
-export function Chat({
-  className,
-  headerTitle = "AutoGPT Copilot",
-  showHeader = true,
-  showSessionInfo = true,
-  showNewChatButton = true,
-  onNewChat,
-  headerActions,
-}: ChatProps) {
-  const {
-    messages,
-    isLoading,
-    isCreating,
-    error,
-    sessionId,
-    createSession,
-    clearSession,
-    loadSession,
-  } = useChat();
-
-  const [isSessionsDrawerOpen, setIsSessionsDrawerOpen] = useState(false);
-
-  const handleNewChat = () => {
-    clearSession();
-    onNewChat?.();
-  };
-
-  const handleSelectSession = async (sessionId: string) => {
-    try {
-      await loadSession(sessionId);
-    } catch (err) {
-      console.error("Failed to load session:", err);
-    }
-  };
-
-  return (
-    <div className={cn("flex h-full flex-col", className)}>
-      {/* Header */}
-      {showHeader && (
-        <header className="shrink-0 border-t border-zinc-200 bg-white p-3">
-          <div className="flex items-center justify-between">
-            <div className="flex items-center gap-3">
-              <button
-                aria-label="View sessions"
-                onClick={() => setIsSessionsDrawerOpen(true)}
-                className="flex size-8 items-center justify-center rounded hover:bg-zinc-100"
-              >
-                <List width="1.25rem" height="1.25rem" />
-              </button>
-              {typeof headerTitle === "string" ? (
-                <Text variant="h2" className="text-lg font-semibold">
-                  {headerTitle}
-                </Text>
-              ) : (
-                headerTitle
-              )}
-            </div>
-            <div className="flex items-center gap-3">
-              {showSessionInfo && sessionId && (
-                <>
-                  {showNewChatButton && (
-                    <Button
-                      variant="outline"
-                      size="small"
-                      onClick={handleNewChat}
-                    >
-                      New Chat
-                    </Button>
-                  )}
-                </>
-              )}
-              {headerActions}
-            </div>
-          </div>
-        </header>
-      )}
-
-      {/* Main Content */}
-      <main className="flex min-h-0 flex-1 flex-col overflow-hidden">
-        {/* Loading State - show when explicitly loading/creating OR when we don't have a session yet and no error */}
-        {(isLoading || isCreating || (!sessionId && !error)) && (
-          <ChatLoadingState
-            message={isCreating ? "Creating session..." : "Loading..."}
-          />
-        )}
-
-        {/* Error State */}
-        {error && !isLoading && (
-          <ChatErrorState error={error} onRetry={createSession} />
-        )}
-
-        {/* Session Content */}
-        {sessionId && !isLoading && !error && (
-          <ChatContainer
-            sessionId={sessionId}
-            initialMessages={messages}
-            className="flex-1"
-          />
-        )}
-      </main>
-
-      {/* Sessions Drawer */}
-      <SessionsDrawer
-        isOpen={isSessionsDrawerOpen}
-        onClose={() => setIsSessionsDrawerOpen(false)}
-        onSelectSession={handleSelectSession}
-        currentSessionId={sessionId}
-      />
-    </div>
-  );
-}

autogpt_platform/frontend/src/app/(platform)/chat/components/Chat/components/ChatContainer/ChatContainer.tsx

@@ -1,88 +0,0 @@
-import type { SessionDetailResponse } from "@/app/api/__generated__/models/sessionDetailResponse";
-import { cn } from "@/lib/utils";
-import { useCallback } from "react";
-import { usePageContext } from "../../usePageContext";
-import { ChatInput } from "../ChatInput/ChatInput";
-import { MessageList } from "../MessageList/MessageList";
-import { QuickActionsWelcome } from "../QuickActionsWelcome/QuickActionsWelcome";
-import { useChatContainer } from "./useChatContainer";
-
-export interface ChatContainerProps {
-  sessionId: string | null;
-  initialMessages: SessionDetailResponse["messages"];
-  className?: string;
-}
-
-export function ChatContainer({
-  sessionId,
-  initialMessages,
-  className,
-}: ChatContainerProps) {
-  const { messages, streamingChunks, isStreaming, sendMessage } =
-    useChatContainer({
-      sessionId,
-      initialMessages,
-    });
-  const { capturePageContext } = usePageContext();
-
-  // Wrap sendMessage to automatically capture page context
-  const sendMessageWithContext = useCallback(
-    async (content: string, isUserMessage: boolean = true) => {
-      const context = capturePageContext();
-      await sendMessage(content, isUserMessage, context);
-    },
-    [sendMessage, capturePageContext],
-  );
-
-  const quickActions = [
-    "Find agents for social media management",
-    "Show me agents for content creation",
-    "Help me automate my business",
-    "What can you help me with?",
-  ];
-
-  return (
-    <div
-      className={cn("flex h-full min-h-0 flex-col", className)}
-      style={{
-        backgroundColor: "#ffffff",
-        backgroundImage:
-          "radial-gradient(#e5e5e5 0.5px, transparent 0.5px), radial-gradient(#e5e5e5 0.5px, #ffffff 0.5px)",
-        backgroundSize: "20px 20px",
-        backgroundPosition: "0 0, 10px 10px",
-      }}
-    >
-      {/* Messages or Welcome Screen */}
-      <div className="flex min-h-0 flex-1 flex-col overflow-hidden pb-24">
-        {messages.length === 0 ? (
-          <QuickActionsWelcome
-            title="Welcome to AutoGPT Copilot"
-            description="Start a conversation to discover and run AI agents."
-            actions={quickActions}
-            onActionClick={sendMessageWithContext}
-            disabled={isStreaming || !sessionId}
-          />
-        ) : (
-          <MessageList
-            messages={messages}
-            streamingChunks={streamingChunks}
-            isStreaming={isStreaming}
-            onSendMessage={sendMessageWithContext}
-            className="flex-1"
-          />
-        )}
-      </div>
-
-      {/* Input - Always visible */}
-      <div className="fixed bottom-0 left-0 right-0 z-50 border-t border-zinc-200 bg-white p-4">
-        <ChatInput
-          onSend={sendMessageWithContext}
-          disabled={isStreaming || !sessionId}
-          placeholder={
-            sessionId ? "Type your message..." : "Creating session..."
-          }
-        />
-      </div>
-    </div>
-  );
-}

autogpt_platform/frontend/src/app/(platform)/chat/components/Chat/components/ChatInput/ChatInput.tsx

@@ -1,64 +0,0 @@
-import { Input } from "@/components/atoms/Input/Input";
-import { cn } from "@/lib/utils";
-import { ArrowUpIcon } from "@phosphor-icons/react";
-import { useChatInput } from "./useChatInput";
-
-export interface ChatInputProps {
-  onSend: (message: string) => void;
-  disabled?: boolean;
-  placeholder?: string;
-  className?: string;
-}
-
-export function ChatInput({
-  onSend,
-  disabled = false,
-  placeholder = "Type your message...",
-  className,
-}: ChatInputProps) {
-  const inputId = "chat-input";
-  const { value, setValue, handleKeyDown, handleSend } = useChatInput({
-    onSend,
-    disabled,
-    maxRows: 5,
-    inputId,
-  });
-
-  return (
-    <div className={cn("relative flex-1", className)}>
-      <Input
-        id={inputId}
-        label="Chat message input"
-        hideLabel
-        type="textarea"
-        value={value}
-        onChange={(e) => setValue(e.target.value)}
-        onKeyDown={handleKeyDown}
-        placeholder={placeholder}
-        disabled={disabled}
-        rows={1}
-        wrapperClassName="mb-0 relative"
-        className="pr-12"
-      />
-      <span id="chat-input-hint" className="sr-only">
-        Press Enter to send, Shift+Enter for new line
-      </span>
-
-      <button
-        onClick={handleSend}
-        disabled={disabled || !value.trim()}
-        className={cn(
-          "absolute right-3 top-1/2 flex h-8 w-8 -translate-y-1/2 items-center justify-center rounded-full",
-          "border border-zinc-800 bg-zinc-800 text-white",
-          "hover:border-zinc-900 hover:bg-zinc-900",
-          "disabled:border-zinc-200 disabled:bg-zinc-200 disabled:text-white disabled:opacity-50",
-          "transition-colors focus-visible:outline-none focus-visible:ring-1 focus-visible:ring-neutral-950",
-          "disabled:pointer-events-none",
-        )}
-        aria-label="Send message"
-      >
-        <ArrowUpIcon className="h-3 w-3" weight="bold" />
-      </button>
-    </div>
-  );
-}

autogpt_platform/frontend/src/app/(platform)/chat/components/Chat/components/ChatInput/useChatInput.ts

@@ -1,60 +0,0 @@
-import { KeyboardEvent, useCallback, useEffect, useState } from "react";
-
-interface UseChatInputArgs {
-  onSend: (message: string) => void;
-  disabled?: boolean;
-  maxRows?: number;
-  inputId?: string;
-}
-
-export function useChatInput({
-  onSend,
-  disabled = false,
-  maxRows = 5,
-  inputId = "chat-input",
-}: UseChatInputArgs) {
-  const [value, setValue] = useState("");
-
-  useEffect(() => {
-    const textarea = document.getElementById(inputId) as HTMLTextAreaElement;
-    if (!textarea) return;
-    textarea.style.height = "auto";
-    const lineHeight = parseInt(
-      window.getComputedStyle(textarea).lineHeight,
-      10,
-    );
-    const maxHeight = lineHeight * maxRows;
-    const newHeight = Math.min(textarea.scrollHeight, maxHeight);
-    textarea.style.height = `${newHeight}px`;
-    textarea.style.overflowY =
-      textarea.scrollHeight > maxHeight ? "auto" : "hidden";
-  }, [value, maxRows, inputId]);
-
-  const handleSend = useCallback(() => {
-    if (disabled || !value.trim()) return;
-    onSend(value.trim());
-    setValue("");
-    const textarea = document.getElementById(inputId) as HTMLTextAreaElement;
-    if (textarea) {
-      textarea.style.height = "auto";
-    }
-  }, [value, onSend, disabled, inputId]);
-
-  const handleKeyDown = useCallback(
-    (event: KeyboardEvent<HTMLInputElement | HTMLTextAreaElement>) => {
-      if (event.key === "Enter" && !event.shiftKey) {
-        event.preventDefault();
-        handleSend();
-      }
-      // Shift+Enter allows default behavior (new line) - no need to handle explicitly
-    },
-    [handleSend],
-  );
-
-  return {
-    value,
-    setValue,
-    handleKeyDown,
-    handleSend,
-  };
-}

autogpt_platform/frontend/src/app/(platform)/chat/components/Chat/components/MessageList/MessageList.tsx

@@ -1,121 +0,0 @@
-"use client";
-
-import { cn } from "@/lib/utils";
-import { ChatMessage } from "../ChatMessage/ChatMessage";
-import type { ChatMessageData } from "../ChatMessage/useChatMessage";
-import { StreamingMessage } from "../StreamingMessage/StreamingMessage";
-import { ThinkingMessage } from "../ThinkingMessage/ThinkingMessage";
-import { useMessageList } from "./useMessageList";
-
-export interface MessageListProps {
-  messages: ChatMessageData[];
-  streamingChunks?: string[];
-  isStreaming?: boolean;
-  className?: string;
-  onStreamComplete?: () => void;
-  onSendMessage?: (content: string) => void;
-}
-
-export function MessageList({
-  messages,
-  streamingChunks = [],
-  isStreaming = false,
-  className,
-  onStreamComplete,
-  onSendMessage,
-}: MessageListProps) {
-  const { messagesEndRef, messagesContainerRef } = useMessageList({
-    messageCount: messages.length,
-    isStreaming,
-  });
-
-  return (
-    <div
-      ref={messagesContainerRef}
-      className={cn(
-        "flex-1 overflow-y-auto",
-        "scrollbar-thin scrollbar-track-transparent scrollbar-thumb-zinc-300",
-        className,
-      )}
-    >
-      <div className="mx-auto flex max-w-3xl flex-col py-4">
-        {/* Render all persisted messages */}
-        {messages.map((message, index) => {
-          // Check if current message is an agent_output tool_response
-          // and if previous message is an assistant message
-          let agentOutput: ChatMessageData | undefined;
-
-          if (message.type === "tool_response" && message.result) {
-            let parsedResult: Record<string, unknown> | null = null;
-            try {
-              parsedResult =
-                typeof message.result === "string"
-                  ? JSON.parse(message.result)
-                  : (message.result as Record<string, unknown>);
-            } catch {
-              parsedResult = null;
-            }
-            if (parsedResult?.type === "agent_output") {
-              const prevMessage = messages[index - 1];
-              if (
-                prevMessage &&
-                prevMessage.type === "message" &&
-                prevMessage.role === "assistant"
-              ) {
-                // This agent output will be rendered inside the previous assistant message
-                // Skip rendering this message separately
-                return null;
-              }
-            }
-          }
-
-          // Check if next message is an agent_output tool_response to include in current assistant message
-          if (message.type === "message" && message.role === "assistant") {
-            const nextMessage = messages[index + 1];
-            if (
-              nextMessage &&
-              nextMessage.type === "tool_response" &&
-              nextMessage.result
-            ) {
-              let parsedResult: Record<string, unknown> | null = null;
-              try {
-                parsedResult =
-                  typeof nextMessage.result === "string"
-                    ? JSON.parse(nextMessage.result)
-                    : (nextMessage.result as Record<string, unknown>);
-              } catch {
-                parsedResult = null;
-              }
-              if (parsedResult?.type === "agent_output") {
-                agentOutput = nextMessage;
-              }
-            }
-          }
-
-          return (
-            <ChatMessage
-              key={index}
-              message={message}
-              onSendMessage={onSendMessage}
-              agentOutput={agentOutput}
-            />
-          );
-        })}
-
-        {/* Render thinking message when streaming but no chunks yet */}
-        {isStreaming && streamingChunks.length === 0 && <ThinkingMessage />}
-
-        {/* Render streaming message if active */}
-        {isStreaming && streamingChunks.length > 0 && (
-          <StreamingMessage
-            chunks={streamingChunks}
-            onComplete={onStreamComplete}
-          />
-        )}
-
-        {/* Invisible div to scroll to */}
-        <div ref={messagesEndRef} />
-      </div>
-    </div>
-  );
-}

autogpt_platform/frontend/src/app/(platform)/chat/components/Chat/components/ToolCallMessage/ToolCallMessage.tsx

@@ -1,24 +0,0 @@
-import { Text } from "@/components/atoms/Text/Text";
-import { cn } from "@/lib/utils";
-import { WrenchIcon } from "@phosphor-icons/react";
-import { getToolActionPhrase } from "../../helpers";
-
-export interface ToolCallMessageProps {
-  toolName: string;
-  className?: string;
-}
-
-export function ToolCallMessage({ toolName, className }: ToolCallMessageProps) {
-  return (
-    <div className={cn("flex items-center justify-center gap-2", className)}>
-      <WrenchIcon
-        size={14}
-        weight="bold"
-        className="flex-shrink-0 text-neutral-500"
-      />
-      <Text variant="small" className="text-neutral-500">
-        {getToolActionPhrase(toolName)}...
-      </Text>
-    </div>
-  );
-}

autogpt_platform/frontend/src/app/(platform)/chat/components/Chat/components/ToolResponseMessage/ToolResponseMessage.tsx

@@ -1,260 +0,0 @@
-import { Text } from "@/components/atoms/Text/Text";
-import "@/components/contextual/OutputRenderers";
-import {
-  globalRegistry,
-  OutputItem,
-} from "@/components/contextual/OutputRenderers";
-import { cn } from "@/lib/utils";
-import type { ToolResult } from "@/types/chat";
-import { WrenchIcon } from "@phosphor-icons/react";
-import { getToolActionPhrase } from "../../helpers";
-
-export interface ToolResponseMessageProps {
-  toolName: string;
-  result?: ToolResult;
-  success?: boolean;
-  className?: string;
-}
-
-export function ToolResponseMessage({
-  toolName,
-  result,
-  success: _success = true,
-  className,
-}: ToolResponseMessageProps) {
-  if (!result) {
-    return (
-      <div className={cn("flex items-center justify-center gap-2", className)}>
-        <WrenchIcon
-          size={14}
-          weight="bold"
-          className="flex-shrink-0 text-neutral-500"
-        />
-        <Text variant="small" className="text-neutral-500">
-          {getToolActionPhrase(toolName)}...
-        </Text>
-      </div>
-    );
-  }
-
-  let parsedResult: Record<string, unknown> | null = null;
-  try {
-    parsedResult =
-      typeof result === "string"
-        ? JSON.parse(result)
-        : (result as Record<string, unknown>);
-  } catch {
-    parsedResult = null;
-  }
-
-  if (parsedResult && typeof parsedResult === "object") {
-    const responseType = parsedResult.type as string | undefined;
-
-    if (responseType === "agent_output") {
-      const execution = parsedResult.execution as
-        | {
-            outputs?: Record<string, unknown[]>;
-          }
-        | null
-        | undefined;
-      const outputs = execution?.outputs || {};
-      const message = parsedResult.message as string | undefined;
-
-      return (
-        <div className={cn("space-y-4 px-4 py-2", className)}>
-          <div className="flex items-center gap-2">
-            <WrenchIcon
-              size={14}
-              weight="bold"
-              className="flex-shrink-0 text-neutral-500"
-            />
-            <Text variant="small" className="text-neutral-500">
-              {getToolActionPhrase(toolName)}
-            </Text>
-          </div>
-          {message && (
-            <div className="rounded border p-4">
-              <Text variant="small" className="text-neutral-600">
-                {message}
-              </Text>
-            </div>
-          )}
-          {Object.keys(outputs).length > 0 && (
-            <div className="space-y-4">
-              {Object.entries(outputs).map(([outputName, values]) =>
-                values.map((value, index) => {
-                  const renderer = globalRegistry.getRenderer(value);
-                  if (renderer) {
-                    return (
-                      <OutputItem
-                        key={`${outputName}-${index}`}
-                        value={value}
-                        renderer={renderer}
-                        label={outputName}
-                      />
-                    );
-                  }
-                  return (
-                    <div
-                      key={`${outputName}-${index}`}
-                      className="rounded border p-4"
-                    >
-                      <Text variant="large-medium" className="mb-2 capitalize">
-                        {outputName}
-                      </Text>
-                      <pre className="overflow-auto text-sm">
-                        {JSON.stringify(value, null, 2)}
-                      </pre>
-                    </div>
-                  );
-                }),
-              )}
-            </div>
-          )}
-        </div>
-      );
-    }
-
-    if (responseType === "block_output" && parsedResult.outputs) {
-      const outputs = parsedResult.outputs as Record<string, unknown[]>;
-
-      return (
-        <div className={cn("space-y-4 px-4 py-2", className)}>
-          <div className="flex items-center gap-2">
-            <WrenchIcon
-              size={14}
-              weight="bold"
-              className="flex-shrink-0 text-neutral-500"
-            />
-            <Text variant="small" className="text-neutral-500">
-              {getToolActionPhrase(toolName)}
-            </Text>
-          </div>
-          <div className="space-y-4">
-            {Object.entries(outputs).map(([outputName, values]) =>
-              values.map((value, index) => {
-                const renderer = globalRegistry.getRenderer(value);
-                if (renderer) {
-                  return (
-                    <OutputItem
-                      key={`${outputName}-${index}`}
-                      value={value}
-                      renderer={renderer}
-                      label={outputName}
-                    />
-                  );
-                }
-                return (
-                  <div
-                    key={`${outputName}-${index}`}
-                    className="rounded border p-4"
-                  >
-                    <Text variant="large-medium" className="mb-2 capitalize">
-                      {outputName}
-                    </Text>
-                    <pre className="overflow-auto text-sm">
-                      {JSON.stringify(value, null, 2)}
-                    </pre>
-                  </div>
-                );
-              }),
-            )}
-          </div>
-        </div>
-      );
-    }
-
-    // Handle other response types with a message field (e.g., understanding_updated)
-    if (parsedResult.message && typeof parsedResult.message === "string") {
-      // Format tool name from snake_case to Title Case
-      const formattedToolName = toolName
-        .split("_")
-        .map((word) => word.charAt(0).toUpperCase() + word.slice(1))
-        .join(" ");
-
-      // Clean up message - remove incomplete user_name references
-      let cleanedMessage = parsedResult.message;
-      // Remove "Updated understanding with: user_name" pattern if user_name is just a placeholder
-      cleanedMessage = cleanedMessage.replace(
-        /Updated understanding with:\s*user_name\.?\s*/gi,
-        "",
-      );
-      // Remove standalone user_name references
-      cleanedMessage = cleanedMessage.replace(/\buser_name\b\.?\s*/gi, "");
-      cleanedMessage = cleanedMessage.trim();
-
-      // Only show message if it has content after cleaning
-      if (!cleanedMessage) {
-        return (
-          <div
-            className={cn(
-              "flex items-center justify-center gap-2 px-4 py-2",
-              className,
-            )}
-          >
-            <WrenchIcon
-              size={14}
-              weight="bold"
-              className="flex-shrink-0 text-neutral-500"
-            />
-            <Text variant="small" className="text-neutral-500">
-              {formattedToolName}
-            </Text>
-          </div>
-        );
-      }
-
-      return (
-        <div className={cn("space-y-2 px-4 py-2", className)}>
-          <div className="flex items-center justify-center gap-2">
-            <WrenchIcon
-              size={14}
-              weight="bold"
-              className="flex-shrink-0 text-neutral-500"
-            />
-            <Text variant="small" className="text-neutral-500">
-              {formattedToolName}
-            </Text>
-          </div>
-          <div className="rounded border p-4">
-            <Text variant="small" className="text-neutral-600">
-              {cleanedMessage}
-            </Text>
-          </div>
-        </div>
-      );
-    }
-  }
-
-  const renderer = globalRegistry.getRenderer(result);
-  if (renderer) {
-    return (
-      <div className={cn("px-4 py-2", className)}>
-        <div className="mb-2 flex items-center gap-2">
-          <WrenchIcon
-            size={14}
-            weight="bold"
-            className="flex-shrink-0 text-neutral-500"
-          />
-          <Text variant="small" className="text-neutral-500">
-            {getToolActionPhrase(toolName)}
-          </Text>
-        </div>
-        <OutputItem value={result} renderer={renderer} />
-      </div>
-    );
-  }
-
-  return (
-    <div className={cn("flex items-center justify-center gap-2", className)}>
-      <WrenchIcon
-        size={14}
-        weight="bold"
-        className="flex-shrink-0 text-neutral-500"
-      />
-      <Text variant="small" className="text-neutral-500">
-        {getToolActionPhrase(toolName)}...
-      </Text>
-    </div>
-  );
-}

autogpt_platform/frontend/src/app/(platform)/chat/components/Chat/helpers.ts

@@ -1,66 +0,0 @@
-/**
- * Maps internal tool names to user-friendly display names with emojis.
- * @deprecated Use getToolActionPhrase or getToolCompletionPhrase for status messages
- *
- * @param toolName - The internal tool name from the backend
- * @returns A user-friendly display name with an emoji prefix
- */
-export function getToolDisplayName(toolName: string): string {
-  const toolDisplayNames: Record<string, string> = {
-    find_agent: "🔍 Search Marketplace",
-    get_agent_details: "📋 Get Agent Details",
-    check_credentials: "🔑 Check Credentials",
-    setup_agent: "⚙️ Setup Agent",
-    run_agent: "▶️ Run Agent",
-    get_required_setup_info: "📝 Get Setup Requirements",
-  };
-  return toolDisplayNames[toolName] || toolName;
-}
-
-/**
- * Maps internal tool names to human-friendly action phrases (present continuous).
- * Used for tool call messages to indicate what action is currently happening.
- *
- * @param toolName - The internal tool name from the backend
- * @returns A human-friendly action phrase in present continuous tense
- */
-export function getToolActionPhrase(toolName: string): string {
-  const toolActionPhrases: Record<string, string> = {
-    find_agent: "Looking for agents in the marketplace",
-    agent_carousel: "Looking for agents in the marketplace",
-    get_agent_details: "Learning about the agent",
-    check_credentials: "Checking your credentials",
-    setup_agent: "Setting up the agent",
-    execution_started: "Running the agent",
-    run_agent: "Running the agent",
-    get_required_setup_info: "Getting setup requirements",
-    schedule_agent: "Scheduling the agent to run",
-  };
-
-  // Return mapped phrase or generate human-friendly fallback
-  return toolActionPhrases[toolName] || toolName;
-}
-
-/**
- * Maps internal tool names to human-friendly completion phrases (past tense).
- * Used for tool response messages to indicate what action was completed.
- *
- * @param toolName - The internal tool name from the backend
- * @returns A human-friendly completion phrase in past tense
- */
-export function getToolCompletionPhrase(toolName: string): string {
-  const toolCompletionPhrases: Record<string, string> = {
-    find_agent: "Finished searching the marketplace",
-    get_agent_details: "Got agent details",
-    check_credentials: "Checked credentials",
-    setup_agent: "Agent setup complete",
-    run_agent: "Agent execution started",
-    get_required_setup_info: "Got setup requirements",
-  };
-
-  // Return mapped phrase or generate human-friendly fallback
-  return (
-    toolCompletionPhrases[toolName] ||
-    `Finished ${toolName.replace(/_/g, " ").replace("...", "")}`
-  );
-}

autogpt_platform/frontend/src/app/(platform)/chat/components/Chat/useChatSession.ts

@@ -1,271 +0,0 @@
-import {
-  getGetV2GetSessionQueryKey,
-  getGetV2GetSessionQueryOptions,
-  postV2CreateSession,
-  useGetV2GetSession,
-  usePatchV2SessionAssignUser,
-  usePostV2CreateSession,
-} from "@/app/api/__generated__/endpoints/chat/chat";
-import type { SessionDetailResponse } from "@/app/api/__generated__/models/sessionDetailResponse";
-import { okData } from "@/app/api/helpers";
-import { isValidUUID } from "@/lib/utils";
-import { Key, storage } from "@/services/storage/local-storage";
-import { useQueryClient } from "@tanstack/react-query";
-import { useCallback, useEffect, useMemo, useRef, useState } from "react";
-import { toast } from "sonner";
-
-interface UseChatSessionArgs {
-  urlSessionId?: string | null;
-  autoCreate?: boolean;
-}
-
-export function useChatSession({
-  urlSessionId,
-  autoCreate = false,
-}: UseChatSessionArgs = {}) {
-  const queryClient = useQueryClient();
-  const [sessionId, setSessionId] = useState<string | null>(null);
-  const [error, setError] = useState<Error | null>(null);
-  const justCreatedSessionIdRef = useRef<string | null>(null);
-
-  useEffect(() => {
-    if (urlSessionId) {
-      if (!isValidUUID(urlSessionId)) {
-        console.error("Invalid session ID format:", urlSessionId);
-        toast.error("Invalid session ID", {
-          description:
-            "The session ID in the URL is not valid. Starting a new session...",
-        });
-        setSessionId(null);
-        storage.clean(Key.CHAT_SESSION_ID);
-        return;
-      }
-      setSessionId(urlSessionId);
-      storage.set(Key.CHAT_SESSION_ID, urlSessionId);
-    } else {
-      const storedSessionId = storage.get(Key.CHAT_SESSION_ID);
-      if (storedSessionId) {
-        if (!isValidUUID(storedSessionId)) {
-          console.error("Invalid stored session ID:", storedSessionId);
-          storage.clean(Key.CHAT_SESSION_ID);
-          setSessionId(null);
-        } else {
-          setSessionId(storedSessionId);
-        }
-      } else if (autoCreate) {
-        setSessionId(null);
-      }
-    }
-  }, [urlSessionId, autoCreate]);
-
-  const {
-    mutateAsync: createSessionMutation,
-    isPending: isCreating,
-    error: createError,
-  } = usePostV2CreateSession();
-
-  const {
-    data: sessionData,
-    isLoading: isLoadingSession,
-    error: loadError,
-    refetch,
-  } = useGetV2GetSession(sessionId || "", {
-    query: {
-      enabled: !!sessionId,
-      select: okData,
-      staleTime: Infinity, // Never mark as stale
-      refetchOnMount: false, // Don't refetch on component mount
-      refetchOnWindowFocus: false, // Don't refetch when window regains focus
-      refetchOnReconnect: false, // Don't refetch when network reconnects
-      retry: 1,
-    },
-  });
-
-  const { mutateAsync: claimSessionMutation } = usePatchV2SessionAssignUser();
-
-  const session = useMemo(() => {
-    if (sessionData) return sessionData;
-
-    if (sessionId && justCreatedSessionIdRef.current === sessionId) {
-      return {
-        id: sessionId,
-        user_id: null,
-        messages: [],
-        created_at: new Date().toISOString(),
-        updated_at: new Date().toISOString(),
-      } as SessionDetailResponse;
-    }
-    return null;
-  }, [sessionData, sessionId]);
-
-  const messages = session?.messages || [];
-  const isLoading = isCreating || isLoadingSession;
-
-  useEffect(() => {
-    if (createError) {
-      setError(
-        createError instanceof Error
-          ? createError
-          : new Error("Failed to create session"),
-      );
-    } else if (loadError) {
-      setError(
-        loadError instanceof Error
-          ? loadError
-          : new Error("Failed to load session"),
-      );
-    } else {
-      setError(null);
-    }
-  }, [createError, loadError]);
-
-  const createSession = useCallback(
-    async function createSession() {
-      try {
-        setError(null);
-        const response = await postV2CreateSession({
-          body: JSON.stringify({}),
-        });
-        if (response.status !== 200) {
-          throw new Error("Failed to create session");
-        }
-        const newSessionId = response.data.id;
-        setSessionId(newSessionId);
-        storage.set(Key.CHAT_SESSION_ID, newSessionId);
-        justCreatedSessionIdRef.current = newSessionId;
-        setTimeout(() => {
-          if (justCreatedSessionIdRef.current === newSessionId) {
-            justCreatedSessionIdRef.current = null;
-          }
-        }, 10000);
-        return newSessionId;
-      } catch (err) {
-        const error =
-          err instanceof Error ? err : new Error("Failed to create session");
-        setError(error);
-        toast.error("Failed to create chat session", {
-          description: error.message,
-        });
-        throw error;
-      }
-    },
-    [createSessionMutation],
-  );
-
-  const loadSession = useCallback(
-    async function loadSession(id: string) {
-      try {
-        setError(null);
-        // Invalidate the query cache for this session to force a fresh fetch
-        await queryClient.invalidateQueries({
-          queryKey: getGetV2GetSessionQueryKey(id),
-        });
-        // Set sessionId after invalidation to ensure the hook refetches
-        setSessionId(id);
-        storage.set(Key.CHAT_SESSION_ID, id);
-        // Force fetch with fresh data (bypass cache)
-        const queryOptions = getGetV2GetSessionQueryOptions(id, {
-          query: {
-            staleTime: 0, // Force fresh fetch
-            retry: 1,
-          },
-        });
-        const result = await queryClient.fetchQuery(queryOptions);
-        if (!result || ("status" in result && result.status !== 200)) {
-          console.warn("Session not found on server, clearing local state");
-          storage.clean(Key.CHAT_SESSION_ID);
-          setSessionId(null);
-          throw new Error("Session not found");
-        }
-      } catch (err) {
-        const error =
-          err instanceof Error ? err : new Error("Failed to load session");
-        setError(error);
-        throw error;
-      }
-    },
-    [queryClient],
-  );
-
-  const refreshSession = useCallback(
-    async function refreshSession() {
-      if (!sessionId) {
-        console.log("[refreshSession] Skipping - no session ID");
-        return;
-      }
-      try {
-        setError(null);
-        await refetch();
-      } catch (err) {
-        const error =
-          err instanceof Error ? err : new Error("Failed to refresh session");
-        setError(error);
-        throw error;
-      }
-    },
-    [sessionId, refetch],
-  );
-
-  const claimSession = useCallback(
-    async function claimSession(id: string) {
-      try {
-        setError(null);
-        await claimSessionMutation({ sessionId: id });
-        if (justCreatedSessionIdRef.current === id) {
-          justCreatedSessionIdRef.current = null;
-        }
-        await queryClient.invalidateQueries({
-          queryKey: getGetV2GetSessionQueryKey(id),
-        });
-        await refetch();
-        toast.success("Session claimed successfully", {
-          description: "Your chat history has been saved to your account",
-        });
-      } catch (err: unknown) {
-        const error =
-          err instanceof Error ? err : new Error("Failed to claim session");
-        const is404 =
-          (typeof err === "object" &&
-            err !== null &&
-            "status" in err &&
-            err.status === 404) ||
-          (typeof err === "object" &&
-            err !== null &&
-            "response" in err &&
-            typeof err.response === "object" &&
-            err.response !== null &&
-            "status" in err.response &&
-            err.response.status === 404);
-        if (!is404) {
-          setError(error);
-          toast.error("Failed to claim session", {
-            description: error.message || "Unable to claim session",
-          });
-        }
-        throw error;
-      }
-    },
-    [claimSessionMutation, queryClient, refetch],
-  );
-
-  const clearSession = useCallback(function clearSession() {
-    setSessionId(null);
-    setError(null);
-    storage.clean(Key.CHAT_SESSION_ID);
-    justCreatedSessionIdRef.current = null;
-  }, []);
-
-  return {
-    session,
-    sessionId,
-    messages,
-    isLoading,
-    isCreating,
-    error,
-    createSession,
-    loadSession,
-    refreshSession,
-    claimSession,
-    clearSession,
-  };
-}

autogpt_platform/frontend/src/app/(platform)/chat/page.tsx

@@ -1,27 +0,0 @@
-"use client";
-
-import { Flag, useGetFlag } from "@/services/feature-flags/use-get-flag";
-import { useRouter } from "next/navigation";
-import { useEffect } from "react";
-import { Chat } from "./components/Chat/Chat";
-
-export default function ChatPage() {
-  const isChatEnabled = useGetFlag(Flag.CHAT);
-  const router = useRouter();
-
-  useEffect(() => {
-    if (isChatEnabled === false) {
-      router.push("/marketplace");
-    }
-  }, [isChatEnabled, router]);
-
-  if (isChatEnabled === null || isChatEnabled === false) {
-    return null;
-  }
-
-  return (
-    <div className="flex h-full flex-col">
-      <Chat className="flex-1" />
-    </div>
-  );
-}

autogpt_platform/frontend/src/app/(platform)/copilot/components/CopilotShell/CopilotShell.tsx

@@ -0,0 +1,88 @@
+"use client";
+
+import { LoadingSpinner } from "@/components/atoms/LoadingSpinner/LoadingSpinner";
+import { NAVBAR_HEIGHT_PX } from "@/lib/constants";
+import type { ReactNode } from "react";
+import { DesktopSidebar } from "./components/DesktopSidebar/DesktopSidebar";
+import { LoadingState } from "./components/LoadingState/LoadingState";
+import { MobileDrawer } from "./components/MobileDrawer/MobileDrawer";
+import { MobileHeader } from "./components/MobileHeader/MobileHeader";
+import { useCopilotShell } from "./useCopilotShell";
+
+interface Props {
+  children: ReactNode;
+}
+
+export function CopilotShell({ children }: Props) {
+  const {
+    isMobile,
+    isDrawerOpen,
+    isLoading,
+    isLoggedIn,
+    hasActiveSession,
+    sessions,
+    currentSessionId,
+    handleSelectSession,
+    handleOpenDrawer,
+    handleCloseDrawer,
+    handleDrawerOpenChange,
+    handleNewChat,
+    hasNextPage,
+    isFetchingNextPage,
+    fetchNextPage,
+    isReadyToShowContent,
+  } = useCopilotShell();
+
+  if (!isLoggedIn) {
+    return (
+      <div className="flex h-full items-center justify-center">
+        <LoadingSpinner size="large" />
+      </div>
+    );
+  }
+
+  return (
+    <div
+      className="flex overflow-hidden bg-[#EFEFF0]"
+      style={{ height: `calc(100vh - ${NAVBAR_HEIGHT_PX}px)` }}
+    >
+      {!isMobile && (
+        <DesktopSidebar
+          sessions={sessions}
+          currentSessionId={currentSessionId}
+          isLoading={isLoading}
+          hasNextPage={hasNextPage}
+          isFetchingNextPage={isFetchingNextPage}
+          onSelectSession={handleSelectSession}
+          onFetchNextPage={fetchNextPage}
+          onNewChat={handleNewChat}
+          hasActiveSession={Boolean(hasActiveSession)}
+        />
+      )}
+
+      <div className="relative flex min-h-0 flex-1 flex-col">
+        {isMobile && <MobileHeader onOpenDrawer={handleOpenDrawer} />}
+        <div className="flex min-h-0 flex-1 flex-col">
+          {isReadyToShowContent ? children : <LoadingState />}
+        </div>
+      </div>
+
+      {isMobile && (
+        <MobileDrawer
+          isOpen={isDrawerOpen}
+          sessions={sessions}
+          currentSessionId={currentSessionId}
+          isLoading={isLoading}
+          hasNextPage={hasNextPage}
+          isFetchingNextPage={isFetchingNextPage}
+          onSelectSession={handleSelectSession}
+          onFetchNextPage={fetchNextPage}
+          onNewChat={handleNewChat}
+          onClose={handleCloseDrawer}
+          onOpenChange={handleDrawerOpenChange}
+          hasActiveSession={Boolean(hasActiveSession)}
+        />
+      )}
+    </div>
+  );
+}