Merge branch 'dev' into claude/fix-block-search-vector-error-QMbO4

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

@@ -31,6 +31,7 @@ class ResponseType(str, Enum):
     # Other
     ERROR = "error"
     USAGE = "usage"
+    HEARTBEAT = "heartbeat"
 
 
 class StreamBaseResponse(BaseModel):
@@ -142,3 +143,20 @@ class StreamError(StreamBaseResponse):
     details: dict[str, Any] | None = Field(
         default=None, description="Additional error details"
     )
+
+
+class StreamHeartbeat(StreamBaseResponse):
+    """Heartbeat to keep SSE connection alive during long-running operations.
+
+    Uses SSE comment format (: comment) which is ignored by clients but keeps
+    the connection alive through proxies and load balancers.
+    """
+
+    type: ResponseType = ResponseType.HEARTBEAT
+    toolCallId: str | None = Field(
+        default=None, description="Tool call ID if heartbeat is for a specific tool"
+    )
+
+    def to_sse(self) -> str:
+        """Convert to SSE comment format to keep connection alive."""
+        return ": heartbeat\n\n"

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

@@ -3,8 +3,6 @@
 import logging
 from typing import Any
 
-from langfuse import observe
-
 from backend.api.features.chat.model import ChatSession
 from backend.data.understanding import (
     BusinessUnderstandingInput,
@@ -61,7 +59,6 @@ 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_output.py

@@ -5,7 +5,6 @@ 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
@@ -329,7 +328,6 @@ 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

@@ -3,8 +3,6 @@
 import logging
 from typing import Any
 
-from langfuse import observe
-
 from backend.api.features.chat.model import ChatSession
 
 from .agent_generator import (
@@ -75,7 +73,6 @@ class CreateAgentTool(BaseTool):
             "required": ["description"],
         }
 
-    @observe(as_type="tool", name="create_agent")
     async def _execute(
         self,
         user_id: str | None,
@@ -116,8 +113,11 @@ class CreateAgentTool(BaseTool):
 
         if decomposition_result is None:
             return ErrorResponse(
-                message="Failed to analyze the goal. Please try rephrasing.",
-                error="Decomposition failed",
+                message="Failed to analyze the goal. The agent generation service may be unavailable or timed out. Please try again.",
+                error="decomposition_failed",
+                details={
+                    "description": description[:100]
+                },  # Include context for debugging
                 session_id=session_id,
             )
 
@@ -182,8 +182,11 @@ class CreateAgentTool(BaseTool):
 
         if agent_json is None:
             return ErrorResponse(
-                message="Failed to generate the agent. Please try again.",
-                error="Generation failed",
+                message="Failed to generate the agent. The agent generation service may be unavailable or timed out. Please try again.",
+                error="generation_failed",
+                details={
+                    "description": description[:100]
+                },  # Include context for debugging
                 session_id=session_id,
             )
 

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

@@ -3,8 +3,6 @@
 import logging
 from typing import Any
 
-from langfuse import observe
-
 from backend.api.features.chat.model import ChatSession
 
 from .agent_generator import (
@@ -81,7 +79,6 @@ class EditAgentTool(BaseTool):
             "required": ["agent_id", "changes"],
         }
 
-    @observe(as_type="tool", name="edit_agent")
     async def _execute(
         self,
         user_id: str | None,
@@ -145,8 +142,9 @@ class EditAgentTool(BaseTool):
 
         if result is None:
             return ErrorResponse(
-                message="Failed to generate changes. Please try rephrasing.",
-                error="Update generation failed",
+                message="Failed to generate changes. The agent generation service may be unavailable or timed out. Please try again.",
+                error="update_generation_failed",
+                details={"agent_id": agent_id, "changes": changes[:100]},
                 session_id=session_id,
             )
 

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

@@ -2,8 +2,6 @@
 
 from typing import Any
 
-from langfuse import observe
-
 from backend.api.features.chat.model import ChatSession
 
 from .agent_search import search_agents
@@ -37,7 +35,6 @@ 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

@@ -1,7 +1,6 @@
 import logging
 from typing import Any
 
-from langfuse import observe
 from prisma.enums import ContentType
 
 from backend.api.features.chat.model import ChatSession
@@ -56,7 +55,6 @@ 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,
@@ -109,7 +107,8 @@ class FindBlockTool(BaseTool):
                 block_id = result["content_id"]
                 block = get_block(block_id)
 
-                if block:
+                # Skip disabled blocks
+                if block and not block.disabled:
                     # Get input/output schemas
                     input_schema = {}
                     output_schema = {}

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

@@ -2,8 +2,6 @@
 
 from typing import Any
 
-from langfuse import observe
-
 from backend.api.features.chat.model import ChatSession
 
 from .agent_search import search_agents
@@ -43,7 +41,6 @@ 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

@@ -4,8 +4,6 @@ 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 (
@@ -73,7 +71,6 @@ 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

@@ -3,7 +3,6 @@
 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
@@ -159,7 +158,6 @@ class RunAgentTool(BaseTool):
         """All operations require authentication."""
         return True
 
-    @observe(as_type="tool", name="run_agent")
     async def _execute(
         self,
         user_id: str | None,

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

@@ -4,8 +4,6 @@ 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
@@ -130,7 +128,6 @@ class RunBlockTool(BaseTool):
 
         return matched_credentials, missing_credentials
 
-    @observe(as_type="tool", name="run_block")
     async def _execute(
         self,
         user_id: str | None,

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

@@ -3,7 +3,6 @@
 import logging
 from typing import Any
 
-from langfuse import observe
 from prisma.enums import ContentType
 
 from backend.api.features.chat.model import ChatSession
@@ -88,7 +87,6 @@ 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/executions/review/review_routes_test.py

@@ -164,9 +164,9 @@ async def test_process_review_action_approve_success(
     """Test successful review approval"""
     # Mock the route functions
 
-    # Mock get_pending_reviews_by_node_exec_ids (called to find the graph_exec_id)
+    # Mock get_reviews_by_node_exec_ids (called to find the graph_exec_id)
     mock_get_reviews_for_user = mocker.patch(
-        "backend.api.features.executions.review.routes.get_pending_reviews_by_node_exec_ids"
+        "backend.api.features.executions.review.routes.get_reviews_by_node_exec_ids"
     )
     mock_get_reviews_for_user.return_value = {"test_node_123": sample_pending_review}
 
@@ -244,9 +244,9 @@ async def test_process_review_action_reject_success(
     """Test successful review rejection"""
     # Mock the route functions
 
-    # Mock get_pending_reviews_by_node_exec_ids (called to find the graph_exec_id)
+    # Mock get_reviews_by_node_exec_ids (called to find the graph_exec_id)
     mock_get_reviews_for_user = mocker.patch(
-        "backend.api.features.executions.review.routes.get_pending_reviews_by_node_exec_ids"
+        "backend.api.features.executions.review.routes.get_reviews_by_node_exec_ids"
     )
     mock_get_reviews_for_user.return_value = {"test_node_123": sample_pending_review}
 
@@ -339,9 +339,9 @@ async def test_process_review_action_mixed_success(
 
     # Mock the route functions
 
-    # Mock get_pending_reviews_by_node_exec_ids (called to find the graph_exec_id)
+    # Mock get_reviews_by_node_exec_ids (called to find the graph_exec_id)
     mock_get_reviews_for_user = mocker.patch(
-        "backend.api.features.executions.review.routes.get_pending_reviews_by_node_exec_ids"
+        "backend.api.features.executions.review.routes.get_reviews_by_node_exec_ids"
     )
     mock_get_reviews_for_user.return_value = {
         "test_node_123": sample_pending_review,
@@ -463,9 +463,9 @@ async def test_process_review_action_review_not_found(
     test_user_id: str,
 ) -> None:
     """Test error when review is not found"""
-    # Mock get_pending_reviews_by_node_exec_ids (called to find the graph_exec_id)
+    # Mock get_reviews_by_node_exec_ids (called to find the graph_exec_id)
     mock_get_reviews_for_user = mocker.patch(
-        "backend.api.features.executions.review.routes.get_pending_reviews_by_node_exec_ids"
+        "backend.api.features.executions.review.routes.get_reviews_by_node_exec_ids"
     )
     # Return empty dict to simulate review not found
     mock_get_reviews_for_user.return_value = {}
@@ -506,7 +506,7 @@ async def test_process_review_action_review_not_found(
     response = await client.post("/api/review/action", json=request_data)
 
     assert response.status_code == 404
-    assert "No pending review found" in response.json()["detail"]
+    assert "Review(s) not found" in response.json()["detail"]
 
 
 @pytest.mark.asyncio(loop_scope="session")
@@ -517,9 +517,9 @@ async def test_process_review_action_partial_failure(
     test_user_id: str,
 ) -> None:
     """Test handling of partial failures in review processing"""
-    # Mock get_pending_reviews_by_node_exec_ids (called to find the graph_exec_id)
+    # Mock get_reviews_by_node_exec_ids (called to find the graph_exec_id)
     mock_get_reviews_for_user = mocker.patch(
-        "backend.api.features.executions.review.routes.get_pending_reviews_by_node_exec_ids"
+        "backend.api.features.executions.review.routes.get_reviews_by_node_exec_ids"
     )
     mock_get_reviews_for_user.return_value = {"test_node_123": sample_pending_review}
 
@@ -567,9 +567,9 @@ async def test_process_review_action_invalid_node_exec_id(
     test_user_id: str,
 ) -> None:
     """Test failure when trying to process review with invalid node execution ID"""
-    # Mock get_pending_reviews_by_node_exec_ids (called to find the graph_exec_id)
+    # Mock get_reviews_by_node_exec_ids (called to find the graph_exec_id)
     mock_get_reviews_for_user = mocker.patch(
-        "backend.api.features.executions.review.routes.get_pending_reviews_by_node_exec_ids"
+        "backend.api.features.executions.review.routes.get_reviews_by_node_exec_ids"
     )
     # Return empty dict to simulate review not found
     mock_get_reviews_for_user.return_value = {}
@@ -596,7 +596,7 @@ async def test_process_review_action_invalid_node_exec_id(
 
     # Returns 404 when review is not found
     assert response.status_code == 404
-    assert "No pending review found" in response.json()["detail"]
+    assert "Review(s) not found" in response.json()["detail"]
 
 
 @pytest.mark.asyncio(loop_scope="session")
@@ -607,9 +607,9 @@ async def test_process_review_action_auto_approve_creates_auto_approval_records(
     test_user_id: str,
 ) -> None:
     """Test that auto_approve_future_actions flag creates auto-approval records"""
-    # Mock get_pending_reviews_by_node_exec_ids (called to find the graph_exec_id)
+    # Mock get_reviews_by_node_exec_ids (called to find the graph_exec_id)
     mock_get_reviews_for_user = mocker.patch(
-        "backend.api.features.executions.review.routes.get_pending_reviews_by_node_exec_ids"
+        "backend.api.features.executions.review.routes.get_reviews_by_node_exec_ids"
     )
     mock_get_reviews_for_user.return_value = {"test_node_123": sample_pending_review}
 
@@ -737,9 +737,9 @@ async def test_process_review_action_without_auto_approve_still_loads_settings(
     test_user_id: str,
 ) -> None:
     """Test that execution context is created with settings even without auto-approve"""
-    # Mock get_pending_reviews_by_node_exec_ids (called to find the graph_exec_id)
+    # Mock get_reviews_by_node_exec_ids (called to find the graph_exec_id)
     mock_get_reviews_for_user = mocker.patch(
-        "backend.api.features.executions.review.routes.get_pending_reviews_by_node_exec_ids"
+        "backend.api.features.executions.review.routes.get_reviews_by_node_exec_ids"
     )
     mock_get_reviews_for_user.return_value = {"test_node_123": sample_pending_review}
 
@@ -885,9 +885,9 @@ async def test_process_review_action_auto_approve_only_applies_to_approved_revie
         reviewed_at=FIXED_NOW,
     )
 
-    # Mock get_pending_reviews_by_node_exec_ids (called to find the graph_exec_id)
+    # Mock get_reviews_by_node_exec_ids (called to find the graph_exec_id)
     mock_get_reviews_for_user = mocker.patch(
-        "backend.api.features.executions.review.routes.get_pending_reviews_by_node_exec_ids"
+        "backend.api.features.executions.review.routes.get_reviews_by_node_exec_ids"
     )
     # Need to return both reviews in WAITING state (before processing)
     approved_review_waiting = PendingHumanReviewModel(
@@ -1031,9 +1031,9 @@ async def test_process_review_action_per_review_auto_approve_granularity(
     test_user_id: str,
 ) -> None:
     """Test that auto-approval can be set per-review (granular control)"""
-    # Mock get_pending_reviews_by_node_exec_ids - return different reviews based on node_exec_id
+    # Mock get_reviews_by_node_exec_ids - return different reviews based on node_exec_id
     mock_get_reviews_for_user = mocker.patch(
-        "backend.api.features.executions.review.routes.get_pending_reviews_by_node_exec_ids"
+        "backend.api.features.executions.review.routes.get_reviews_by_node_exec_ids"
     )
 
     # Create a mapping of node_exec_id to review

autogpt_platform/backend/backend/api/features/executions/review/routes.py

@@ -14,9 +14,9 @@ from backend.data.execution import (
 from backend.data.graph import get_graph_settings
 from backend.data.human_review import (
     create_auto_approval_record,
-    get_pending_reviews_by_node_exec_ids,
     get_pending_reviews_for_execution,
     get_pending_reviews_for_user,
+    get_reviews_by_node_exec_ids,
     has_pending_reviews_for_graph_exec,
     process_all_reviews_for_execution,
 )
@@ -137,17 +137,17 @@ async def process_review_action(
             detail="At least one review must be provided",
         )
 
-    # Batch fetch all requested reviews
-    reviews_map = await get_pending_reviews_by_node_exec_ids(
+    # Batch fetch all requested reviews (regardless of status for idempotent handling)
+    reviews_map = await get_reviews_by_node_exec_ids(
         list(all_request_node_ids), user_id
     )
 
-    # Validate all reviews were found
+    # Validate all reviews were found (must exist, any status is OK for now)
     missing_ids = all_request_node_ids - set(reviews_map.keys())
     if missing_ids:
         raise HTTPException(
             status_code=status.HTTP_404_NOT_FOUND,
-            detail=f"No pending review found for node execution(s): {', '.join(missing_ids)}",
+            detail=f"Review(s) not found: {', '.join(missing_ids)}",
         )
 
     # Validate all reviews belong to the same execution

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

@@ -188,6 +188,10 @@ class BlockHandler(ContentHandler):
             try:
                 block_instance = block_cls()
 
+                # Skip disabled blocks - they shouldn't be indexed
+                if block_instance.disabled:
+                    continue
+
                 # Build searchable text from block metadata
                 parts = []
                 if hasattr(block_instance, "name") and block_instance.name:
@@ -248,12 +252,19 @@ class BlockHandler(ContentHandler):
         from backend.data.block import get_blocks
 
         all_blocks = get_blocks()
-        total_blocks = len(all_blocks)
+
+        # Filter out disabled blocks - they're not indexed
+        enabled_block_ids = [
+            block_id
+            for block_id, block_cls in all_blocks.items()
+            if not block_cls().disabled
+        ]
+        total_blocks = len(enabled_block_ids)
 
         if total_blocks == 0:
             return {"total": 0, "with_embeddings": 0, "without_embeddings": 0}
 
-        block_ids = list(all_blocks.keys())
+        block_ids = enabled_block_ids
         placeholders = ",".join([f"${i+1}" for i in range(len(block_ids))])
 
         embedded_result = await query_raw_with_schema(

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

@@ -81,6 +81,7 @@ async def test_block_handler_get_missing_items(mocker):
     mock_block_instance.name = "Calculator Block"
     mock_block_instance.description = "Performs calculations"
     mock_block_instance.categories = [MagicMock(value="MATH")]
+    mock_block_instance.disabled = False
     mock_block_instance.input_schema.model_json_schema.return_value = {
         "properties": {"expression": {"description": "Math expression to evaluate"}}
     }
@@ -116,11 +117,18 @@ async def test_block_handler_get_stats(mocker):
     """Test BlockHandler returns correct stats."""
     handler = BlockHandler()
 
-    # Mock get_blocks
+    # Mock get_blocks - each block class returns an instance with disabled=False
+    def make_mock_block_class():
+        mock_class = MagicMock()
+        mock_instance = MagicMock()
+        mock_instance.disabled = False
+        mock_class.return_value = mock_instance
+        return mock_class
+
     mock_blocks = {
-        "block-1": MagicMock(),
-        "block-2": MagicMock(),
-        "block-3": MagicMock(),
+        "block-1": make_mock_block_class(),
+        "block-2": make_mock_block_class(),
+        "block-3": make_mock_block_class(),
     }
 
     # Mock embedded count query (2 blocks have embeddings)
@@ -309,6 +317,7 @@ async def test_block_handler_handles_missing_attributes():
     mock_block_class = MagicMock()
     mock_block_instance = MagicMock()
     mock_block_instance.name = "Minimal Block"
+    mock_block_instance.disabled = False
     # No description, categories, or schema
     del mock_block_instance.description
     del mock_block_instance.categories
@@ -342,6 +351,7 @@ async def test_block_handler_skips_failed_blocks():
     good_instance.name = "Good Block"
     good_instance.description = "Works fine"
     good_instance.categories = []
+    good_instance.disabled = False
     good_block.return_value = good_instance
 
     bad_block = MagicMock()

autogpt_platform/backend/backend/data/human_review.py

@@ -263,11 +263,14 @@ async def get_pending_review_by_node_exec_id(
     return PendingHumanReviewModel.from_db(review, node_id=node_id)
 
 
-async def get_pending_reviews_by_node_exec_ids(
+async def get_reviews_by_node_exec_ids(
     node_exec_ids: list[str], user_id: str
 ) -> dict[str, "PendingHumanReviewModel"]:
     """
-    Get multiple pending reviews by their node execution IDs in a single batch query.
+    Get multiple reviews by their node execution IDs regardless of status.
+
+    Unlike get_pending_reviews_by_node_exec_ids, this returns reviews in any status
+    (WAITING, APPROVED, REJECTED). Used for validation in idempotent operations.
 
     Args:
         node_exec_ids: List of node execution IDs to look up
@@ -283,7 +286,6 @@ async def get_pending_reviews_by_node_exec_ids(
         where={
             "nodeExecId": {"in": node_exec_ids},
             "userId": user_id,
-            "status": ReviewStatus.WAITING,
         }
     )
 
@@ -407,38 +409,68 @@ async def process_all_reviews_for_execution(
 ) -> dict[str, PendingHumanReviewModel]:
     """Process all pending reviews for an execution with approve/reject decisions.
 
+    Handles race conditions gracefully: if a review was already processed with the
+    same decision by a concurrent request, it's treated as success rather than error.
+
     Args:
         user_id: User ID for ownership validation
         review_decisions: Map of node_exec_id -> (status, reviewed_data, message)
 
     Returns:
-        Dict of node_exec_id -> updated review model
+        Dict of node_exec_id -> updated review model (includes already-processed reviews)
     """
     if not review_decisions:
         return {}
 
     node_exec_ids = list(review_decisions.keys())
 
-    # Get all reviews for validation
-    reviews = await PendingHumanReview.prisma().find_many(
+    # Get all reviews (both WAITING and already processed) for the user
+    all_reviews = await PendingHumanReview.prisma().find_many(
         where={
             "nodeExecId": {"in": node_exec_ids},
             "userId": user_id,
-            "status": ReviewStatus.WAITING,
         },
     )
 
-    # Validate all reviews can be processed
-    if len(reviews) != len(node_exec_ids):
-        missing_ids = set(node_exec_ids) - {review.nodeExecId for review in reviews}
+    # Separate into pending and already-processed reviews
+    reviews_to_process = []
+    already_processed = []
+    for review in all_reviews:
+        if review.status == ReviewStatus.WAITING:
+            reviews_to_process.append(review)
+        else:
+            already_processed.append(review)
+
+    # Check for truly missing reviews (not found at all)
+    found_ids = {review.nodeExecId for review in all_reviews}
+    missing_ids = set(node_exec_ids) - found_ids
+    if missing_ids:
         raise ValueError(
-            f"Reviews not found, access denied, or not in WAITING status: {', '.join(missing_ids)}"
+            f"Reviews not found or access denied: {', '.join(missing_ids)}"
         )
 
-    # Create parallel update tasks
+    # Validate already-processed reviews have compatible status (same decision)
+    # This handles race conditions where another request processed the same reviews
+    for review in already_processed:
+        requested_status = review_decisions[review.nodeExecId][0]
+        if review.status != requested_status:
+            raise ValueError(
+                f"Review {review.nodeExecId} was already processed with status "
+                f"{review.status}, cannot change to {requested_status}"
+            )
+
+    # Log if we're handling a race condition (some reviews already processed)
+    if already_processed:
+        already_processed_ids = [r.nodeExecId for r in already_processed]
+        logger.info(
+            f"Race condition handled: {len(already_processed)} review(s) already "
+            f"processed by concurrent request: {already_processed_ids}"
+        )
+
+    # Create parallel update tasks for reviews that still need processing
     update_tasks = []
 
-    for review in reviews:
+    for review in reviews_to_process:
         new_status, reviewed_data, message = review_decisions[review.nodeExecId]
         has_data_changes = reviewed_data is not None and reviewed_data != review.payload
 
@@ -463,7 +495,7 @@ async def process_all_reviews_for_execution(
         update_tasks.append(task)
 
     # Execute all updates in parallel and get updated reviews
-    updated_reviews = await asyncio.gather(*update_tasks)
+    updated_reviews = await asyncio.gather(*update_tasks) if update_tasks else []
 
     # Note: Execution resumption is now handled at the API layer after ALL reviews
     # for an execution are processed (both approved and rejected)
@@ -472,8 +504,11 @@ async def process_all_reviews_for_execution(
     # Local import to avoid event loop conflicts in tests
     from backend.data.execution import get_node_execution
 
+    # Combine updated reviews with already-processed ones (for idempotent response)
+    all_result_reviews = list(updated_reviews) + already_processed
+
     result = {}
-    for review in updated_reviews:
+    for review in all_result_reviews:
         node_exec = await get_node_execution(review.nodeExecId)
         node_id = node_exec.node_id if node_exec else review.nodeExecId
         result[review.nodeExecId] = PendingHumanReviewModel.from_db(

autogpt_platform/frontend/src/app/(platform)/copilot/NewChatContext.tsx

@@ -1,41 +0,0 @@
-"use client";
-
-import { createContext, useContext, useRef, type ReactNode } from "react";
-
-interface NewChatContextValue {
-  onNewChatClick: () => void;
-  setOnNewChatClick: (handler?: () => void) => void;
-  performNewChat?: () => void;
-  setPerformNewChat: (handler?: () => void) => void;
-}
-
-const NewChatContext = createContext<NewChatContextValue | null>(null);
-
-export function NewChatProvider({ children }: { children: ReactNode }) {
-  const onNewChatRef = useRef<(() => void) | undefined>();
-  const performNewChatRef = useRef<(() => void) | undefined>();
-  const contextValueRef = useRef<NewChatContextValue>({
-    onNewChatClick() {
-      onNewChatRef.current?.();
-    },
-    setOnNewChatClick(handler?: () => void) {
-      onNewChatRef.current = handler;
-    },
-    performNewChat() {
-      performNewChatRef.current?.();
-    },
-    setPerformNewChat(handler?: () => void) {
-      performNewChatRef.current = handler;
-    },
-  });
-
-  return (
-    <NewChatContext.Provider value={contextValueRef.current}>
-      {children}
-    </NewChatContext.Provider>
-  );
-}
-
-export function useNewChat() {
-  return useContext(NewChatContext);
-}

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

@@ -4,7 +4,7 @@ import { ChatLoader } from "@/components/contextual/Chat/components/ChatLoader/C
 import { NAVBAR_HEIGHT_PX } from "@/lib/constants";
 import type { ReactNode } from "react";
 import { useEffect } from "react";
-import { useNewChat } from "../../NewChatContext";
+import { useCopilotStore } from "../../copilot-page-store";
 import { DesktopSidebar } from "./components/DesktopSidebar/DesktopSidebar";
 import { LoadingState } from "./components/LoadingState/LoadingState";
 import { MobileDrawer } from "./components/MobileDrawer/MobileDrawer";
@@ -35,21 +35,23 @@ export function CopilotShell({ children }: Props) {
     isReadyToShowContent,
   } = useCopilotShell();
 
-  const newChatContext = useNewChat();
-  const handleNewChatClickWrapper =
-    newChatContext?.onNewChatClick || handleNewChat;
+  const setNewChatHandler = useCopilotStore((s) => s.setNewChatHandler);
+  const requestNewChat = useCopilotStore((s) => s.requestNewChat);
 
   useEffect(
     function registerNewChatHandler() {
-      if (!newChatContext) return;
-      newChatContext.setPerformNewChat(handleNewChat);
+      setNewChatHandler(handleNewChat);
       return function cleanup() {
-        newChatContext.setPerformNewChat(undefined);
+        setNewChatHandler(null);
       };
     },
-    [newChatContext, handleNewChat],
+    [handleNewChat],
   );
 
+  function handleNewChatClick() {
+    requestNewChat();
+  }
+
   if (!isLoggedIn) {
     return (
       <div className="flex h-full items-center justify-center">
@@ -72,7 +74,7 @@ export function CopilotShell({ children }: Props) {
           isFetchingNextPage={isFetchingNextPage}
           onSelectSession={handleSelectSession}
           onFetchNextPage={fetchNextPage}
-          onNewChat={handleNewChatClickWrapper}
+          onNewChat={handleNewChatClick}
           hasActiveSession={Boolean(hasActiveSession)}
         />
       )}
@@ -94,7 +96,7 @@ export function CopilotShell({ children }: Props) {
           isFetchingNextPage={isFetchingNextPage}
           onSelectSession={handleSelectSession}
           onFetchNextPage={fetchNextPage}
-          onNewChat={handleNewChatClickWrapper}
+          onNewChat={handleNewChatClick}
           onClose={handleCloseDrawer}
           onOpenChange={handleDrawerOpenChange}
           hasActiveSession={Boolean(hasActiveSession)}

autogpt_platform/frontend/src/app/(platform)/copilot/components/CopilotShell/components/SessionsList/useSessionsPagination.ts

@@ -1,7 +1,12 @@
-import { useGetV2ListSessions } from "@/app/api/__generated__/endpoints/chat/chat";
+import {
+  getGetV2ListSessionsQueryKey,
+  useGetV2ListSessions,
+} from "@/app/api/__generated__/endpoints/chat/chat";
 import type { SessionSummaryResponse } from "@/app/api/__generated__/models/sessionSummaryResponse";
 import { okData } from "@/app/api/helpers";
-import { useEffect, useMemo, useState } from "react";
+import { useChatStore } from "@/components/contextual/Chat/chat-store";
+import { useQueryClient } from "@tanstack/react-query";
+import { useEffect, useState } from "react";
 
 const PAGE_SIZE = 50;
 
@@ -15,6 +20,8 @@ export function useSessionsPagination({ enabled }: UseSessionsPaginationArgs) {
     SessionSummaryResponse[]
   >([]);
   const [totalCount, setTotalCount] = useState<number | null>(null);
+  const queryClient = useQueryClient();
+  const onStreamComplete = useChatStore((state) => state.onStreamComplete);
 
   const { data, isLoading, isFetching, isError } = useGetV2ListSessions(
     { limit: PAGE_SIZE, offset },
@@ -25,35 +32,47 @@ export function useSessionsPagination({ enabled }: UseSessionsPaginationArgs) {
     },
   );
 
-  useEffect(() => {
-    const responseData = okData(data);
-    if (responseData) {
-      const newSessions = responseData.sessions;
-      const total = responseData.total;
-      setTotalCount(total);
-
-      if (offset === 0) {
-        setAccumulatedSessions(newSessions);
-      } else {
-        setAccumulatedSessions((prev) => [...prev, ...newSessions]);
-      }
-    } else if (!enabled) {
+  useEffect(function refreshOnStreamComplete() {
+    const unsubscribe = onStreamComplete(function handleStreamComplete() {
+      setOffset(0);
       setAccumulatedSessions([]);
       setTotalCount(null);
-    }
-  }, [data, offset, enabled]);
+      queryClient.invalidateQueries({
+        queryKey: getGetV2ListSessionsQueryKey(),
+      });
+    });
+    return unsubscribe;
+  }, []);
 
-  const hasNextPage = useMemo(() => {
-    if (totalCount === null) return false;
-    return accumulatedSessions.length < totalCount;
-  }, [accumulatedSessions.length, totalCount]);
+  useEffect(
+    function updateSessionsFromResponse() {
+      const responseData = okData(data);
+      if (responseData) {
+        const newSessions = responseData.sessions;
+        const total = responseData.total;
+        setTotalCount(total);
 
-  const areAllSessionsLoaded = useMemo(() => {
-    if (totalCount === null) return false;
-    return (
-      accumulatedSessions.length >= totalCount && !isFetching && !isLoading
-    );
-  }, [accumulatedSessions.length, totalCount, isFetching, isLoading]);
+        if (offset === 0) {
+          setAccumulatedSessions(newSessions);
+        } else {
+          setAccumulatedSessions((prev) => [...prev, ...newSessions]);
+        }
+      } else if (!enabled) {
+        setAccumulatedSessions([]);
+        setTotalCount(null);
+      }
+    },
+    [data, offset, enabled],
+  );
+
+  const hasNextPage =
+    totalCount !== null && accumulatedSessions.length < totalCount;
+
+  const areAllSessionsLoaded =
+    totalCount !== null &&
+    accumulatedSessions.length >= totalCount &&
+    !isFetching &&
+    !isLoading;
 
   useEffect(() => {
     if (

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

@@ -2,9 +2,7 @@ import type { SessionDetailResponse } from "@/app/api/__generated__/models/sessi
 import type { SessionSummaryResponse } from "@/app/api/__generated__/models/sessionSummaryResponse";
 import { format, formatDistanceToNow, isToday } from "date-fns";
 
-export function convertSessionDetailToSummary(
-  session: SessionDetailResponse,
-): SessionSummaryResponse {
+export function convertSessionDetailToSummary(session: SessionDetailResponse) {
   return {
     id: session.id,
     created_at: session.created_at,
@@ -13,17 +11,25 @@ export function convertSessionDetailToSummary(
   };
 }
 
-export function filterVisibleSessions(
-  sessions: SessionSummaryResponse[],
-): SessionSummaryResponse[] {
-  return sessions.filter(
-    (session) => session.updated_at !== session.created_at,
-  );
+export function filterVisibleSessions(sessions: SessionSummaryResponse[]) {
+  const fiveMinutesAgo = Date.now() - 5 * 60 * 1000;
+  return sessions.filter((session) => {
+    const hasBeenUpdated = session.updated_at !== session.created_at;
+
+    if (hasBeenUpdated) return true;
+
+    const isRecentlyCreated =
+      new Date(session.created_at).getTime() > fiveMinutesAgo;
+
+    return isRecentlyCreated;
+  });
 }
 
-export function getSessionTitle(session: SessionSummaryResponse): string {
+export function getSessionTitle(session: SessionSummaryResponse) {
   if (session.title) return session.title;
+
   const isNewSession = session.updated_at === session.created_at;
+
   if (isNewSession) {
     const createdDate = new Date(session.created_at);
     if (isToday(createdDate)) {
@@ -31,12 +37,11 @@ export function getSessionTitle(session: SessionSummaryResponse): string {
     }
     return format(createdDate, "MMM d, yyyy");
   }
+
   return "Untitled Chat";
 }
 
-export function getSessionUpdatedLabel(
-  session: SessionSummaryResponse,
-): string {
+export function getSessionUpdatedLabel(session: SessionSummaryResponse) {
   if (!session.updated_at) return "";
   return formatDistanceToNow(new Date(session.updated_at), { addSuffix: true });
 }
@@ -45,8 +50,10 @@ export function mergeCurrentSessionIntoList(
   accumulatedSessions: SessionSummaryResponse[],
   currentSessionId: string | null,
   currentSessionData: SessionDetailResponse | null | undefined,
-): SessionSummaryResponse[] {
+  recentlyCreatedSessions?: Map<string, SessionSummaryResponse>,
+) {
   const filteredSessions: SessionSummaryResponse[] = [];
+  const addedIds = new Set<string>();
 
   if (accumulatedSessions.length > 0) {
     const visibleSessions = filterVisibleSessions(accumulatedSessions);
@@ -61,29 +68,40 @@ export function mergeCurrentSessionIntoList(
         );
         if (!isInVisible) {
           filteredSessions.push(currentInAll);
+          addedIds.add(currentInAll.id);
         }
       }
     }
 
-    filteredSessions.push(...visibleSessions);
+    for (const session of visibleSessions) {
+      if (!addedIds.has(session.id)) {
+        filteredSessions.push(session);
+        addedIds.add(session.id);
+      }
+    }
   }
 
   if (currentSessionId && currentSessionData) {
-    const isCurrentInList = filteredSessions.some(
-      (s) => s.id === currentSessionId,
-    );
-    if (!isCurrentInList) {
+    if (!addedIds.has(currentSessionId)) {
       const summarySession = convertSessionDetailToSummary(currentSessionData);
       filteredSessions.unshift(summarySession);
+      addedIds.add(currentSessionId);
+    }
+  }
+
+  if (recentlyCreatedSessions) {
+    for (const [sessionId, sessionData] of recentlyCreatedSessions) {
+      if (!addedIds.has(sessionId)) {
+        filteredSessions.unshift(sessionData);
+        addedIds.add(sessionId);
+      }
     }
   }
 
   return filteredSessions;
 }
 
-export function getCurrentSessionId(
-  searchParams: URLSearchParams,
-): string | null {
+export function getCurrentSessionId(searchParams: URLSearchParams) {
   return searchParams.get("sessionId");
 }
 
@@ -95,11 +113,7 @@ export function shouldAutoSelectSession(
   accumulatedSessions: SessionSummaryResponse[],
   isLoading: boolean,
   totalCount: number | null,
-): {
-  shouldSelect: boolean;
-  sessionIdToSelect: string | null;
-  shouldCreate: boolean;
-} {
+) {
   if (!areAllSessionsLoaded || hasAutoSelectedSession) {
     return {
       shouldSelect: false,
@@ -146,7 +160,7 @@ export function checkReadyToShowContent(
   isCurrentSessionLoading: boolean,
   currentSessionData: SessionDetailResponse | null | undefined,
   hasAutoSelectedSession: boolean,
-): boolean {
+) {
   if (!areAllSessionsLoaded) return false;
 
   if (paramSessionId) {

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

@@ -4,23 +4,25 @@ import {
   getGetV2ListSessionsQueryKey,
   useGetV2GetSession,
 } from "@/app/api/__generated__/endpoints/chat/chat";
+import type { SessionSummaryResponse } from "@/app/api/__generated__/models/sessionSummaryResponse";
 import { okData } from "@/app/api/helpers";
 import { useBreakpoint } from "@/lib/hooks/useBreakpoint";
 import { useSupabase } from "@/lib/supabase/hooks/useSupabase";
 import { useQueryClient } from "@tanstack/react-query";
-import { usePathname, useRouter, useSearchParams } from "next/navigation";
+import { parseAsString, useQueryState } from "nuqs";
+import { usePathname, useSearchParams } from "next/navigation";
 import { useEffect, useRef, useState } from "react";
 import { useMobileDrawer } from "./components/MobileDrawer/useMobileDrawer";
 import { useSessionsPagination } from "./components/SessionsList/useSessionsPagination";
 import {
   checkReadyToShowContent,
+  convertSessionDetailToSummary,
   filterVisibleSessions,
   getCurrentSessionId,
   mergeCurrentSessionIntoList,
 } from "./helpers";
 
 export function useCopilotShell() {
-  const router = useRouter();
   const pathname = usePathname();
   const searchParams = useSearchParams();
   const queryClient = useQueryClient();
@@ -29,6 +31,8 @@ export function useCopilotShell() {
   const isMobile =
     breakpoint === "base" || breakpoint === "sm" || breakpoint === "md";
 
+  const [, setUrlSessionId] = useQueryState("sessionId", parseAsString);
+
   const isOnHomepage = pathname === "/copilot";
   const paramSessionId = searchParams.get("sessionId");
 
@@ -65,6 +69,9 @@ export function useCopilotShell() {
 
   const [hasAutoSelectedSession, setHasAutoSelectedSession] = useState(false);
   const hasAutoSelectedRef = useRef(false);
+  const recentlyCreatedSessionsRef = useRef<
+    Map<string, SessionSummaryResponse>
+  >(new Map());
 
   // Mark as auto-selected when sessionId is in URL
   useEffect(() => {
@@ -91,6 +98,30 @@ export function useCopilotShell() {
     }
   }, [isOnHomepage, paramSessionId, queryClient]);
 
+  // Track newly created sessions to ensure they stay visible even when switching away
+  useEffect(() => {
+    if (currentSessionId && currentSessionData) {
+      const isNewSession =
+        currentSessionData.updated_at === currentSessionData.created_at;
+      const isNotInAccumulated = !accumulatedSessions.some(
+        (s) => s.id === currentSessionId,
+      );
+      if (isNewSession || isNotInAccumulated) {
+        const summary = convertSessionDetailToSummary(currentSessionData);
+        recentlyCreatedSessionsRef.current.set(currentSessionId, summary);
+      }
+    }
+  }, [currentSessionId, currentSessionData, accumulatedSessions]);
+
+  // Clean up recently created sessions that are now in the accumulated list
+  useEffect(() => {
+    for (const sessionId of recentlyCreatedSessionsRef.current.keys()) {
+      if (accumulatedSessions.some((s) => s.id === sessionId)) {
+        recentlyCreatedSessionsRef.current.delete(sessionId);
+      }
+    }
+  }, [accumulatedSessions]);
+
   // Reset pagination when query becomes disabled
   const prevPaginationEnabledRef = useRef(paginationEnabled);
   useEffect(() => {
@@ -105,6 +136,7 @@ export function useCopilotShell() {
     accumulatedSessions,
     currentSessionId,
     currentSessionData,
+    recentlyCreatedSessionsRef.current,
   );
 
   const visibleSessions = filterVisibleSessions(sessions);
@@ -124,22 +156,17 @@ export function useCopilotShell() {
       );
 
   function handleSelectSession(sessionId: string) {
-    // Navigate using replaceState to avoid full page reload
-    window.history.replaceState(null, "", `/copilot?sessionId=${sessionId}`);
-    // Force a re-render by updating the URL through router
-    router.replace(`/copilot?sessionId=${sessionId}`);
+    setUrlSessionId(sessionId, { shallow: false });
     if (isMobile) handleCloseDrawer();
   }
 
   function handleNewChat() {
     resetAutoSelect();
     resetPagination();
-    // Invalidate and refetch sessions list to ensure newly created sessions appear
     queryClient.invalidateQueries({
       queryKey: getGetV2ListSessionsQueryKey(),
     });
-    window.history.replaceState(null, "", "/copilot");
-    router.replace("/copilot");
+    setUrlSessionId(null, { shallow: false });
     if (isMobile) handleCloseDrawer();
   }
 

autogpt_platform/frontend/src/app/(platform)/copilot/copilot-page-store.ts

@@ -0,0 +1,54 @@
+"use client";
+
+import { create } from "zustand";
+
+interface CopilotStoreState {
+  isStreaming: boolean;
+  isNewChatModalOpen: boolean;
+  newChatHandler: (() => void) | null;
+}
+
+interface CopilotStoreActions {
+  setIsStreaming: (isStreaming: boolean) => void;
+  setNewChatHandler: (handler: (() => void) | null) => void;
+  requestNewChat: () => void;
+  confirmNewChat: () => void;
+  cancelNewChat: () => void;
+}
+
+type CopilotStore = CopilotStoreState & CopilotStoreActions;
+
+export const useCopilotStore = create<CopilotStore>((set, get) => ({
+  isStreaming: false,
+  isNewChatModalOpen: false,
+  newChatHandler: null,
+
+  setIsStreaming(isStreaming) {
+    set({ isStreaming });
+  },
+
+  setNewChatHandler(handler) {
+    set({ newChatHandler: handler });
+  },
+
+  requestNewChat() {
+    const { isStreaming, newChatHandler } = get();
+    if (isStreaming) {
+      set({ isNewChatModalOpen: true });
+    } else if (newChatHandler) {
+      newChatHandler();
+    }
+  },
+
+  confirmNewChat() {
+    const { newChatHandler } = get();
+    set({ isNewChatModalOpen: false });
+    if (newChatHandler) {
+      newChatHandler();
+    }
+  },
+
+  cancelNewChat() {
+    set({ isNewChatModalOpen: false });
+  },
+}));

autogpt_platform/frontend/src/app/(platform)/copilot/layout.tsx

@@ -1,11 +1,6 @@
 import type { ReactNode } from "react";
-import { NewChatProvider } from "./NewChatContext";
 import { CopilotShell } from "./components/CopilotShell/CopilotShell";
 
 export default function CopilotLayout({ children }: { children: ReactNode }) {
-  return (
-    <NewChatProvider>
-      <CopilotShell>{children}</CopilotShell>
-    </NewChatProvider>
-  );
+  return <CopilotShell>{children}</CopilotShell>;
 }

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

@@ -1,16 +1,19 @@
 "use client";
 
-import { Skeleton } from "@/components/__legacy__/ui/skeleton";
 import { Button } from "@/components/atoms/Button/Button";
+
+import { Skeleton } from "@/components/atoms/Skeleton/Skeleton";
 import { Text } from "@/components/atoms/Text/Text";
 import { Chat } from "@/components/contextual/Chat/Chat";
 import { ChatInput } from "@/components/contextual/Chat/components/ChatInput/ChatInput";
 import { ChatLoader } from "@/components/contextual/Chat/components/ChatLoader/ChatLoader";
 import { Dialog } from "@/components/molecules/Dialog/Dialog";
+import { useCopilotStore } from "./copilot-page-store";
 import { useCopilotPage } from "./useCopilotPage";
 
 export default function CopilotPage() {
   const { state, handlers } = useCopilotPage();
+  const confirmNewChat = useCopilotStore((s) => s.confirmNewChat);
   const {
     greetingName,
     quickActions,
@@ -25,15 +28,11 @@ export default function CopilotPage() {
     handleSessionNotFound,
     handleStreamingChange,
     handleCancelNewChat,
-    proceedWithNewChat,
     handleNewChatModalOpen,
   } = handlers;
 
-  if (!isReady) {
-    return null;
-  }
+  if (!isReady) return null;
 
-  // Show Chat when we have an active session
   if (pageState.type === "chat") {
     return (
       <div className="flex h-full flex-col">
@@ -71,7 +70,7 @@ export default function CopilotPage() {
                 <Button
                   type="button"
                   variant="primary"
-                  onClick={proceedWithNewChat}
+                  onClick={confirmNewChat}
                 >
                   Start new chat
                 </Button>
@@ -83,7 +82,7 @@ export default function CopilotPage() {
     );
   }
 
-  if (pageState.type === "newChat") {
+  if (pageState.type === "newChat" || pageState.type === "creating") {
     return (
       <div className="flex h-full flex-1 flex-col items-center justify-center bg-[#f8f8f9]">
         <div className="flex flex-col items-center gap-4">
@@ -96,21 +95,6 @@ export default function CopilotPage() {
     );
   }
 
-  // Show loading state while creating session and sending first message
-  if (pageState.type === "creating") {
-    return (
-      <div className="flex h-full flex-1 flex-col items-center justify-center bg-[#f8f8f9]">
-        <div className="flex flex-col items-center gap-4">
-          <ChatLoader />
-          <Text variant="body" className="text-zinc-500">
-            Loading your chats...
-          </Text>
-        </div>
-      </div>
-    );
-  }
-
-  // Show Welcome screen
   return (
     <div className="flex h-full flex-1 items-center justify-center overflow-y-auto bg-[#f8f8f9] px-6 py-10">
       <div className="w-full text-center">

autogpt_platform/frontend/src/app/(platform)/copilot/useCopilotPage.ts

@@ -1,4 +1,7 @@
-import { postV2CreateSession } from "@/app/api/__generated__/endpoints/chat/chat";
+import {
+  getGetV2ListSessionsQueryKey,
+  postV2CreateSession,
+} from "@/app/api/__generated__/endpoints/chat/chat";
 import { useToast } from "@/components/molecules/Toast/use-toast";
 import { getHomepageRoute } from "@/lib/constants";
 import { useSupabase } from "@/lib/supabase/hooks/useSupabase";
@@ -8,25 +11,22 @@ import {
   useGetFlag,
 } from "@/services/feature-flags/use-get-flag";
 import * as Sentry from "@sentry/nextjs";
+import { useQueryClient } from "@tanstack/react-query";
 import { useFlags } from "launchdarkly-react-client-sdk";
 import { useRouter } from "next/navigation";
 import { useEffect, useReducer } from "react";
-import { useNewChat } from "./NewChatContext";
+import { useCopilotStore } from "./copilot-page-store";
 import { getGreetingName, getQuickActions, type PageState } from "./helpers";
 import { useCopilotURLState } from "./useCopilotURLState";
 
 type CopilotState = {
   pageState: PageState;
-  isStreaming: boolean;
-  isNewChatModalOpen: boolean;
   initialPrompts: Record<string, string>;
   previousSessionId: string | null;
 };
 
 type CopilotAction =
   | { type: "setPageState"; pageState: PageState }
-  | { type: "setStreaming"; isStreaming: boolean }
-  | { type: "setNewChatModalOpen"; isOpen: boolean }
   | { type: "setInitialPrompt"; sessionId: string; prompt: string }
   | { type: "setPreviousSessionId"; sessionId: string | null };
 
@@ -52,14 +52,6 @@ function copilotReducer(
     if (isSamePageState(action.pageState, state.pageState)) return state;
     return { ...state, pageState: action.pageState };
   }
-  if (action.type === "setStreaming") {
-    if (action.isStreaming === state.isStreaming) return state;
-    return { ...state, isStreaming: action.isStreaming };
-  }
-  if (action.type === "setNewChatModalOpen") {
-    if (action.isOpen === state.isNewChatModalOpen) return state;
-    return { ...state, isNewChatModalOpen: action.isOpen };
-  }
   if (action.type === "setInitialPrompt") {
     if (state.initialPrompts[action.sessionId] === action.prompt) return state;
     return {
@@ -79,9 +71,14 @@ function copilotReducer(
 
 export function useCopilotPage() {
   const router = useRouter();
+  const queryClient = useQueryClient();
   const { user, isLoggedIn, isUserLoading } = useSupabase();
   const { toast } = useToast();
 
+  const isNewChatModalOpen = useCopilotStore((s) => s.isNewChatModalOpen);
+  const setIsStreaming = useCopilotStore((s) => s.setIsStreaming);
+  const cancelNewChat = useCopilotStore((s) => s.cancelNewChat);
+
   const isChatEnabled = useGetFlag(Flag.CHAT);
   const flags = useFlags<FlagValues>();
   const homepageRoute = getHomepageRoute(isChatEnabled);
@@ -93,13 +90,10 @@ export function useCopilotPage() {
 
   const [state, dispatch] = useReducer(copilotReducer, {
     pageState: { type: "welcome" },
-    isStreaming: false,
-    isNewChatModalOpen: false,
     initialPrompts: {},
     previousSessionId: null,
   });
 
-  const newChatContext = useNewChat();
   const greetingName = getGreetingName(user);
   const quickActions = getQuickActions();
 
@@ -124,17 +118,6 @@ export function useCopilotPage() {
     setPreviousSessionId,
   });
 
-  useEffect(
-    function registerNewChatHandler() {
-      if (!newChatContext) return;
-      newChatContext.setOnNewChatClick(handleNewChatClick);
-      return function cleanup() {
-        newChatContext.setOnNewChatClick(undefined);
-      };
-    },
-    [newChatContext, handleNewChatClick],
-  );
-
   useEffect(
     function transitionNewChatToWelcome() {
       if (state.pageState.type === "newChat") {
@@ -189,6 +172,10 @@ export function useCopilotPage() {
         prompt: trimmedPrompt,
       });
 
+      await queryClient.invalidateQueries({
+        queryKey: getGetV2ListSessionsQueryKey(),
+      });
+
       await setUrlSessionId(sessionId, { shallow: false });
       dispatch({
         type: "setPageState",
@@ -211,37 +198,15 @@ export function useCopilotPage() {
   }
 
   function handleStreamingChange(isStreamingValue: boolean) {
-    dispatch({ type: "setStreaming", isStreaming: isStreamingValue });
-  }
-
-  async function proceedWithNewChat() {
-    dispatch({ type: "setNewChatModalOpen", isOpen: false });
-    if (newChatContext?.performNewChat) {
-      newChatContext.performNewChat();
-      return;
-    }
-    try {
-      await setUrlSessionId(null, { shallow: false });
-    } catch (error) {
-      console.error("[CopilotPage] Failed to clear session:", error);
-    }
-    router.replace("/copilot");
+    setIsStreaming(isStreamingValue);
   }
 
   function handleCancelNewChat() {
-    dispatch({ type: "setNewChatModalOpen", isOpen: false });
+    cancelNewChat();
   }
 
   function handleNewChatModalOpen(isOpen: boolean) {
-    dispatch({ type: "setNewChatModalOpen", isOpen });
-  }
-
-  function handleNewChatClick() {
-    if (state.isStreaming) {
-      dispatch({ type: "setNewChatModalOpen", isOpen: true });
-    } else {
-      proceedWithNewChat();
-    }
+    if (!isOpen) cancelNewChat();
   }
 
   return {
@@ -250,7 +215,7 @@ export function useCopilotPage() {
       quickActions,
       isLoading: isUserLoading,
       pageState: state.pageState,
-      isNewChatModalOpen: state.isNewChatModalOpen,
+      isNewChatModalOpen,
       isReady: isFlagReady && isChatEnabled !== false && isLoggedIn,
     },
     handlers: {
@@ -259,7 +224,6 @@ export function useCopilotPage() {
       handleSessionNotFound,
       handleStreamingChange,
       handleCancelNewChat,
-      proceedWithNewChat,
       handleNewChatModalOpen,
     },
   };

autogpt_platform/frontend/src/app/(platform)/layout.tsx

@@ -1,10 +1,12 @@
 import { Navbar } from "@/components/layout/Navbar/Navbar";
+import { NetworkStatusMonitor } from "@/services/network-status/NetworkStatusMonitor";
 import { ReactNode } from "react";
 import { AdminImpersonationBanner } from "./admin/components/AdminImpersonationBanner";
 
 export default function PlatformLayout({ children }: { children: ReactNode }) {
   return (
     <main className="flex h-screen w-full flex-col">
+      <NetworkStatusMonitor />
       <Navbar />
       <AdminImpersonationBanner />
       <section className="flex-1">{children}</section>

autogpt_platform/frontend/src/components/atoms/Skeleton/Skeleton.tsx

@@ -0,0 +1,14 @@
+import { cn } from "@/lib/utils";
+
+interface Props extends React.HTMLAttributes<HTMLDivElement> {
+  className?: string;
+}
+
+export function Skeleton({ className, ...props }: Props) {
+  return (
+    <div
+      className={cn("animate-pulse rounded-md bg-zinc-100", className)}
+      {...props}
+    />
+  );
+}

autogpt_platform/frontend/src/components/atoms/Skeleton/skeleton.stories.tsx

@@ -1,4 +1,4 @@
-import { Skeleton } from "@/components/__legacy__/ui/skeleton";
+import { Skeleton } from "./Skeleton";
 import type { Meta, StoryObj } from "@storybook/nextjs";
 
 const meta: Meta<typeof Skeleton> = {

autogpt_platform/frontend/src/components/contextual/Chat/chat-store.ts

@@ -0,0 +1,234 @@
+"use client";
+
+import { create } from "zustand";
+import type {
+  ActiveStream,
+  StreamChunk,
+  StreamCompleteCallback,
+  StreamResult,
+  StreamStatus,
+} from "./chat-types";
+import { executeStream } from "./stream-executor";
+
+const COMPLETED_STREAM_TTL = 5 * 60 * 1000; // 5 minutes
+
+interface ChatStoreState {
+  activeStreams: Map<string, ActiveStream>;
+  completedStreams: Map<string, StreamResult>;
+  activeSessions: Set<string>;
+  streamCompleteCallbacks: Set<StreamCompleteCallback>;
+}
+
+interface ChatStoreActions {
+  startStream: (
+    sessionId: string,
+    message: string,
+    isUserMessage: boolean,
+    context?: { url: string; content: string },
+    onChunk?: (chunk: StreamChunk) => void,
+  ) => Promise<void>;
+  stopStream: (sessionId: string) => void;
+  subscribeToStream: (
+    sessionId: string,
+    onChunk: (chunk: StreamChunk) => void,
+    skipReplay?: boolean,
+  ) => () => void;
+  getStreamStatus: (sessionId: string) => StreamStatus;
+  getCompletedStream: (sessionId: string) => StreamResult | undefined;
+  clearCompletedStream: (sessionId: string) => void;
+  isStreaming: (sessionId: string) => boolean;
+  registerActiveSession: (sessionId: string) => void;
+  unregisterActiveSession: (sessionId: string) => void;
+  isSessionActive: (sessionId: string) => boolean;
+  onStreamComplete: (callback: StreamCompleteCallback) => () => void;
+}
+
+type ChatStore = ChatStoreState & ChatStoreActions;
+
+function notifyStreamComplete(
+  callbacks: Set<StreamCompleteCallback>,
+  sessionId: string,
+) {
+  for (const callback of callbacks) {
+    try {
+      callback(sessionId);
+    } catch (err) {
+      console.warn("[ChatStore] Stream complete callback error:", err);
+    }
+  }
+}
+
+function cleanupCompletedStreams(completedStreams: Map<string, StreamResult>) {
+  const now = Date.now();
+  for (const [sessionId, result] of completedStreams) {
+    if (now - result.completedAt > COMPLETED_STREAM_TTL) {
+      completedStreams.delete(sessionId);
+    }
+  }
+}
+
+function moveToCompleted(
+  activeStreams: Map<string, ActiveStream>,
+  completedStreams: Map<string, StreamResult>,
+  streamCompleteCallbacks: Set<StreamCompleteCallback>,
+  sessionId: string,
+) {
+  const stream = activeStreams.get(sessionId);
+  if (!stream) return;
+
+  const result: StreamResult = {
+    sessionId,
+    status: stream.status,
+    chunks: stream.chunks,
+    completedAt: Date.now(),
+    error: stream.error,
+  };
+
+  completedStreams.set(sessionId, result);
+  activeStreams.delete(sessionId);
+  cleanupCompletedStreams(completedStreams);
+
+  if (stream.status === "completed" || stream.status === "error") {
+    notifyStreamComplete(streamCompleteCallbacks, sessionId);
+  }
+}
+
+export const useChatStore = create<ChatStore>((set, get) => ({
+  activeStreams: new Map(),
+  completedStreams: new Map(),
+  activeSessions: new Set(),
+  streamCompleteCallbacks: new Set(),
+
+  startStream: async function startStream(
+    sessionId,
+    message,
+    isUserMessage,
+    context,
+    onChunk,
+  ) {
+    const { activeStreams, completedStreams, streamCompleteCallbacks } = get();
+
+    const existingStream = activeStreams.get(sessionId);
+    if (existingStream) {
+      existingStream.abortController.abort();
+      moveToCompleted(
+        activeStreams,
+        completedStreams,
+        streamCompleteCallbacks,
+        sessionId,
+      );
+    }
+
+    const abortController = new AbortController();
+    const initialCallbacks = new Set<(chunk: StreamChunk) => void>();
+    if (onChunk) initialCallbacks.add(onChunk);
+
+    const stream: ActiveStream = {
+      sessionId,
+      abortController,
+      status: "streaming",
+      startedAt: Date.now(),
+      chunks: [],
+      onChunkCallbacks: initialCallbacks,
+    };
+
+    activeStreams.set(sessionId, stream);
+
+    try {
+      await executeStream(stream, message, isUserMessage, context);
+    } finally {
+      if (onChunk) stream.onChunkCallbacks.delete(onChunk);
+      if (stream.status !== "streaming") {
+        moveToCompleted(
+          activeStreams,
+          completedStreams,
+          streamCompleteCallbacks,
+          sessionId,
+        );
+      }
+    }
+  },
+
+  stopStream: function stopStream(sessionId) {
+    const { activeStreams, completedStreams, streamCompleteCallbacks } = get();
+    const stream = activeStreams.get(sessionId);
+    if (stream) {
+      stream.abortController.abort();
+      stream.status = "completed";
+      moveToCompleted(
+        activeStreams,
+        completedStreams,
+        streamCompleteCallbacks,
+        sessionId,
+      );
+    }
+  },
+
+  subscribeToStream: function subscribeToStream(
+    sessionId,
+    onChunk,
+    skipReplay = false,
+  ) {
+    const { activeStreams } = get();
+
+    const stream = activeStreams.get(sessionId);
+    if (stream) {
+      if (!skipReplay) {
+        for (const chunk of stream.chunks) {
+          onChunk(chunk);
+        }
+      }
+      stream.onChunkCallbacks.add(onChunk);
+      return function unsubscribe() {
+        stream.onChunkCallbacks.delete(onChunk);
+      };
+    }
+
+    return function noop() {};
+  },
+
+  getStreamStatus: function getStreamStatus(sessionId) {
+    const { activeStreams, completedStreams } = get();
+
+    const active = activeStreams.get(sessionId);
+    if (active) return active.status;
+
+    const completed = completedStreams.get(sessionId);
+    if (completed) return completed.status;
+
+    return "idle";
+  },
+
+  getCompletedStream: function getCompletedStream(sessionId) {
+    return get().completedStreams.get(sessionId);
+  },
+
+  clearCompletedStream: function clearCompletedStream(sessionId) {
+    get().completedStreams.delete(sessionId);
+  },
+
+  isStreaming: function isStreaming(sessionId) {
+    const stream = get().activeStreams.get(sessionId);
+    return stream?.status === "streaming";
+  },
+
+  registerActiveSession: function registerActiveSession(sessionId) {
+    get().activeSessions.add(sessionId);
+  },
+
+  unregisterActiveSession: function unregisterActiveSession(sessionId) {
+    get().activeSessions.delete(sessionId);
+  },
+
+  isSessionActive: function isSessionActive(sessionId) {
+    return get().activeSessions.has(sessionId);
+  },
+
+  onStreamComplete: function onStreamComplete(callback) {
+    const { streamCompleteCallbacks } = get();
+    streamCompleteCallbacks.add(callback);
+    return function unsubscribe() {
+      streamCompleteCallbacks.delete(callback);
+    };
+  },
+}));

autogpt_platform/frontend/src/components/contextual/Chat/chat-types.ts

@@ -0,0 +1,94 @@
+import type { ToolArguments, ToolResult } from "@/types/chat";
+
+export type StreamStatus = "idle" | "streaming" | "completed" | "error";
+
+export interface StreamChunk {
+  type:
+    | "text_chunk"
+    | "text_ended"
+    | "tool_call"
+    | "tool_call_start"
+    | "tool_response"
+    | "login_needed"
+    | "need_login"
+    | "credentials_needed"
+    | "error"
+    | "usage"
+    | "stream_end";
+  timestamp?: string;
+  content?: string;
+  message?: string;
+  code?: string;
+  details?: Record<string, unknown>;
+  tool_id?: string;
+  tool_name?: string;
+  arguments?: ToolArguments;
+  result?: ToolResult;
+  success?: boolean;
+  idx?: number;
+  session_id?: string;
+  agent_info?: {
+    graph_id: string;
+    name: string;
+    trigger_type: string;
+  };
+  provider?: string;
+  provider_name?: string;
+  credential_type?: string;
+  scopes?: string[];
+  title?: string;
+  [key: string]: unknown;
+}
+
+export type VercelStreamChunk =
+  | { type: "start"; messageId: string }
+  | { type: "finish" }
+  | { type: "text-start"; id: string }
+  | { type: "text-delta"; id: string; delta: string }
+  | { type: "text-end"; id: string }
+  | { type: "tool-input-start"; toolCallId: string; toolName: string }
+  | {
+      type: "tool-input-available";
+      toolCallId: string;
+      toolName: string;
+      input: Record<string, unknown>;
+    }
+  | {
+      type: "tool-output-available";
+      toolCallId: string;
+      toolName?: string;
+      output: unknown;
+      success?: boolean;
+    }
+  | {
+      type: "usage";
+      promptTokens: number;
+      completionTokens: number;
+      totalTokens: number;
+    }
+  | {
+      type: "error";
+      errorText: string;
+      code?: string;
+      details?: Record<string, unknown>;
+    };
+
+export interface ActiveStream {
+  sessionId: string;
+  abortController: AbortController;
+  status: StreamStatus;
+  startedAt: number;
+  chunks: StreamChunk[];
+  error?: Error;
+  onChunkCallbacks: Set<(chunk: StreamChunk) => void>;
+}
+
+export interface StreamResult {
+  sessionId: string;
+  status: StreamStatus;
+  chunks: StreamChunk[];
+  completedAt: number;
+  error?: Error;
+}
+
+export type StreamCompleteCallback = (sessionId: string) => void;

autogpt_platform/frontend/src/components/contextual/Chat/components/ChatContainer/ChatContainer.tsx

@@ -4,6 +4,7 @@ import { Text } from "@/components/atoms/Text/Text";
 import { Dialog } from "@/components/molecules/Dialog/Dialog";
 import { useBreakpoint } from "@/lib/hooks/useBreakpoint";
 import { cn } from "@/lib/utils";
+import { GlobeHemisphereEastIcon } from "@phosphor-icons/react";
 import { useEffect } from "react";
 import { ChatInput } from "../ChatInput/ChatInput";
 import { MessageList } from "../MessageList/MessageList";
@@ -55,24 +56,37 @@ export function ChatContainer({
       )}
     >
       <Dialog
-        title="Service unavailable"
+        title={
+          <div className="flex items-center gap-2">
+            <GlobeHemisphereEastIcon className="size-6" />
+            <Text
+              variant="body"
+              className="text-md font-poppins leading-none md:text-lg"
+            >
+              Service unavailable
+            </Text>
+          </div>
+        }
         controlled={{
           isOpen: isRegionBlockedModalOpen,
           set: handleRegionModalOpenChange,
         }}
         onClose={handleRegionModalClose}
+        styling={{ maxWidth: 550, width: "100%", minWidth: "auto" }}
       >
         <Dialog.Content>
-          <div className="flex flex-col gap-4">
+          <div className="flex flex-col gap-8">
             <Text variant="body">
-              This model is not available in your region. Please connect via VPN
-              and try again.
+              The Autogpt AI model is not available in your region or your
+              connection is blocking it. Please try again with a different
+              connection.
             </Text>
-            <div className="flex justify-end">
+            <div className="flex justify-center">
               <Button
                 type="button"
                 variant="primary"
                 onClick={handleRegionModalClose}
+                className="w-full"
               >
                 Got it
               </Button>

autogpt_platform/frontend/src/components/contextual/Chat/components/ChatContainer/createStreamEventDispatcher.ts

@@ -1,5 +1,5 @@
 import { toast } from "sonner";
-import { StreamChunk } from "../../useChatStream";
+import type { StreamChunk } from "../../chat-types";
 import type { HandlerDependencies } from "./handlers";
 import {
   handleError,

autogpt_platform/frontend/src/components/contextual/Chat/components/ChatContainer/helpers.ts

@@ -1,7 +1,118 @@
+import type { SessionDetailResponse } from "@/app/api/__generated__/models/sessionDetailResponse";
 import { SessionKey, sessionStorage } from "@/services/storage/session-storage";
 import type { ToolResult } from "@/types/chat";
 import type { ChatMessageData } from "../ChatMessage/useChatMessage";
 
+export function processInitialMessages(
+  initialMessages: SessionDetailResponse["messages"],
+): ChatMessageData[] {
+  const processedMessages: ChatMessageData[] = [];
+  const toolCallMap = new Map<string, string>();
+
+  for (const msg of initialMessages) {
+    if (!isValidMessage(msg)) {
+      console.warn("Invalid message structure from backend:", msg);
+      continue;
+    }
+
+    let content = String(msg.content || "");
+    const role = String(msg.role || "assistant").toLowerCase();
+    const toolCalls = msg.tool_calls;
+    const timestamp = msg.timestamp
+      ? new Date(msg.timestamp as string)
+      : undefined;
+
+    if (role === "user") {
+      content = removePageContext(content);
+      if (!content.trim()) continue;
+      processedMessages.push({
+        type: "message",
+        role: "user",
+        content,
+        timestamp,
+      });
+      continue;
+    }
+
+    if (role === "assistant") {
+      content = content
+        .replace(/<thinking>[\s\S]*?<\/thinking>/gi, "")
+        .replace(/<internal_reasoning>[\s\S]*?<\/internal_reasoning>/gi, "")
+        .trim();
+
+      if (toolCalls && isToolCallArray(toolCalls) && toolCalls.length > 0) {
+        for (const toolCall of toolCalls) {
+          const toolName = toolCall.function.name;
+          const toolId = toolCall.id;
+          toolCallMap.set(toolId, toolName);
+
+          try {
+            const args = JSON.parse(toolCall.function.arguments || "{}");
+            processedMessages.push({
+              type: "tool_call",
+              toolId,
+              toolName,
+              arguments: args,
+              timestamp,
+            });
+          } catch (err) {
+            console.warn("Failed to parse tool call arguments:", err);
+            processedMessages.push({
+              type: "tool_call",
+              toolId,
+              toolName,
+              arguments: {},
+              timestamp,
+            });
+          }
+        }
+        if (content.trim()) {
+          processedMessages.push({
+            type: "message",
+            role: "assistant",
+            content,
+            timestamp,
+          });
+        }
+      } else if (content.trim()) {
+        processedMessages.push({
+          type: "message",
+          role: "assistant",
+          content,
+          timestamp,
+        });
+      }
+      continue;
+    }
+
+    if (role === "tool") {
+      const toolCallId = (msg.tool_call_id as string) || "";
+      const toolName = toolCallMap.get(toolCallId) || "unknown";
+      const toolResponse = parseToolResponse(
+        content,
+        toolCallId,
+        toolName,
+        timestamp,
+      );
+      if (toolResponse) {
+        processedMessages.push(toolResponse);
+      }
+      continue;
+    }
+
+    if (content.trim()) {
+      processedMessages.push({
+        type: "message",
+        role: role as "user" | "assistant" | "system",
+        content,
+        timestamp,
+      });
+    }
+  }
+
+  return processedMessages;
+}
+
 export function hasSentInitialPrompt(sessionId: string): boolean {
   try {
     const sent = JSON.parse(

autogpt_platform/frontend/src/components/contextual/Chat/components/ChatContainer/useChatContainer.ts

@@ -1,5 +1,6 @@
 import type { SessionDetailResponse } from "@/app/api/__generated__/models/sessionDetailResponse";
-import { useCallback, useEffect, useMemo, useRef, useState } from "react";
+import { useEffect, useMemo, useRef, useState } from "react";
+import { useChatStore } from "../../chat-store";
 import { toast } from "sonner";
 import { useChatStream } from "../../useChatStream";
 import { usePageContext } from "../../usePageContext";
@@ -9,11 +10,8 @@ import {
   createUserMessage,
   filterAuthMessages,
   hasSentInitialPrompt,
-  isToolCallArray,
-  isValidMessage,
   markInitialPromptSent,
-  parseToolResponse,
-  removePageContext,
+  processInitialMessages,
 } from "./helpers";
 
 interface Args {
@@ -41,11 +39,18 @@ export function useChatContainer({
     sendMessage: sendStreamMessage,
     stopStreaming,
   } = useChatStream();
+  const activeStreams = useChatStore((s) => s.activeStreams);
+  const subscribeToStream = useChatStore((s) => s.subscribeToStream);
   const isStreaming = isStreamingInitiated || hasTextChunks;
 
-  useEffect(() => {
-    if (sessionId !== previousSessionIdRef.current) {
-      stopStreaming(previousSessionIdRef.current ?? undefined, true);
+  useEffect(
+    function handleSessionChange() {
+      if (sessionId === previousSessionIdRef.current) return;
+
+      const prevSession = previousSessionIdRef.current;
+      if (prevSession) {
+        stopStreaming(prevSession);
+      }
       previousSessionIdRef.current = sessionId;
       setMessages([]);
       setStreamingChunks([]);
@@ -53,138 +58,11 @@ export function useChatContainer({
       setHasTextChunks(false);
       setIsStreamingInitiated(false);
       hasResponseRef.current = false;
-    }
-  }, [sessionId, stopStreaming]);
 
-  const allMessages = useMemo(() => {
-    const processedInitialMessages: ChatMessageData[] = [];
-    const toolCallMap = new Map<string, string>();
+      if (!sessionId) return;
 
-    for (const msg of initialMessages) {
-      if (!isValidMessage(msg)) {
-        console.warn("Invalid message structure from backend:", msg);
-        continue;
-      }
-
-      let content = String(msg.content || "");
-      const role = String(msg.role || "assistant").toLowerCase();
-      const toolCalls = msg.tool_calls;
-      const timestamp = msg.timestamp
-        ? new Date(msg.timestamp as string)
-        : undefined;
-
-      if (role === "user") {
-        content = removePageContext(content);
-        if (!content.trim()) continue;
-        processedInitialMessages.push({
-          type: "message",
-          role: "user",
-          content,
-          timestamp,
-        });
-        continue;
-      }
-
-      if (role === "assistant") {
-        content = content
-          .replace(/<thinking>[\s\S]*?<\/thinking>/gi, "")
-          .trim();
-
-        if (toolCalls && isToolCallArray(toolCalls) && toolCalls.length > 0) {
-          for (const toolCall of toolCalls) {
-            const toolName = toolCall.function.name;
-            const toolId = toolCall.id;
-            toolCallMap.set(toolId, toolName);
-
-            try {
-              const args = JSON.parse(toolCall.function.arguments || "{}");
-              processedInitialMessages.push({
-                type: "tool_call",
-                toolId,
-                toolName,
-                arguments: args,
-                timestamp,
-              });
-            } catch (err) {
-              console.warn("Failed to parse tool call arguments:", err);
-              processedInitialMessages.push({
-                type: "tool_call",
-                toolId,
-                toolName,
-                arguments: {},
-                timestamp,
-              });
-            }
-          }
-          if (content.trim()) {
-            processedInitialMessages.push({
-              type: "message",
-              role: "assistant",
-              content,
-              timestamp,
-            });
-          }
-        } else if (content.trim()) {
-          processedInitialMessages.push({
-            type: "message",
-            role: "assistant",
-            content,
-            timestamp,
-          });
-        }
-        continue;
-      }
-
-      if (role === "tool") {
-        const toolCallId = (msg.tool_call_id as string) || "";
-        const toolName = toolCallMap.get(toolCallId) || "unknown";
-        const toolResponse = parseToolResponse(
-          content,
-          toolCallId,
-          toolName,
-          timestamp,
-        );
-        if (toolResponse) {
-          processedInitialMessages.push(toolResponse);
-        }
-        continue;
-      }
-
-      if (content.trim()) {
-        processedInitialMessages.push({
-          type: "message",
-          role: role as "user" | "assistant" | "system",
-          content,
-          timestamp,
-        });
-      }
-    }
-
-    return [...processedInitialMessages, ...messages];
-  }, [initialMessages, messages]);
-
-  const sendMessage = useCallback(
-    async function sendMessage(
-      content: string,
-      isUserMessage: boolean = true,
-      context?: { url: string; content: string },
-    ) {
-      if (!sessionId) {
-        console.error("[useChatContainer] Cannot send message: no session ID");
-        return;
-      }
-      setIsRegionBlockedModalOpen(false);
-      if (isUserMessage) {
-        const userMessage = createUserMessage(content);
-        setMessages((prev) => [...filterAuthMessages(prev), userMessage]);
-      } else {
-        setMessages((prev) => filterAuthMessages(prev));
-      }
-      setStreamingChunks([]);
-      streamingChunksRef.current = [];
-      setHasTextChunks(false);
-      setIsStreamingInitiated(true);
-      hasResponseRef.current = false;
+      const activeStream = activeStreams.get(sessionId);
+      if (!activeStream || activeStream.status !== "streaming") return;
 
       const dispatcher = createStreamEventDispatcher({
         setHasTextChunks,
@@ -197,42 +75,85 @@ export function useChatContainer({
         setIsStreamingInitiated,
       });
 
-      try {
-        await sendStreamMessage(
-          sessionId,
-          content,
-          dispatcher,
-          isUserMessage,
-          context,
-        );
-      } catch (err) {
-        console.error("[useChatContainer] Failed to send message:", err);
-        setIsStreamingInitiated(false);
-
-        // Don't show error toast for AbortError (expected during cleanup)
-        if (err instanceof Error && err.name === "AbortError") return;
-
-        const errorMessage =
-          err instanceof Error ? err.message : "Failed to send message";
-        toast.error("Failed to send message", {
-          description: errorMessage,
-        });
-      }
+      setIsStreamingInitiated(true);
+      const skipReplay = initialMessages.length > 0;
+      return subscribeToStream(sessionId, dispatcher, skipReplay);
     },
-    [sessionId, sendStreamMessage],
+    [sessionId, stopStreaming, activeStreams, subscribeToStream],
   );
 
-  const handleStopStreaming = useCallback(() => {
+  const allMessages = useMemo(
+    () => [...processInitialMessages(initialMessages), ...messages],
+    [initialMessages, messages],
+  );
+
+  async function sendMessage(
+    content: string,
+    isUserMessage: boolean = true,
+    context?: { url: string; content: string },
+  ) {
+    if (!sessionId) {
+      console.error("[useChatContainer] Cannot send message: no session ID");
+      return;
+    }
+    setIsRegionBlockedModalOpen(false);
+    if (isUserMessage) {
+      const userMessage = createUserMessage(content);
+      setMessages((prev) => [...filterAuthMessages(prev), userMessage]);
+    } else {
+      setMessages((prev) => filterAuthMessages(prev));
+    }
+    setStreamingChunks([]);
+    streamingChunksRef.current = [];
+    setHasTextChunks(false);
+    setIsStreamingInitiated(true);
+    hasResponseRef.current = false;
+
+    const dispatcher = createStreamEventDispatcher({
+      setHasTextChunks,
+      setStreamingChunks,
+      streamingChunksRef,
+      hasResponseRef,
+      setMessages,
+      setIsRegionBlockedModalOpen,
+      sessionId,
+      setIsStreamingInitiated,
+    });
+
+    try {
+      await sendStreamMessage(
+        sessionId,
+        content,
+        dispatcher,
+        isUserMessage,
+        context,
+      );
+    } catch (err) {
+      console.error("[useChatContainer] Failed to send message:", err);
+      setIsStreamingInitiated(false);
+
+      if (err instanceof Error && err.name === "AbortError") return;
+
+      const errorMessage =
+        err instanceof Error ? err.message : "Failed to send message";
+      toast.error("Failed to send message", {
+        description: errorMessage,
+      });
+    }
+  }
+
+  function handleStopStreaming() {
     stopStreaming();
     setStreamingChunks([]);
     streamingChunksRef.current = [];
     setHasTextChunks(false);
     setIsStreamingInitiated(false);
-  }, [stopStreaming]);
+  }
 
   const { capturePageContext } = usePageContext();
+  const sendMessageRef = useRef(sendMessage);
+  sendMessageRef.current = sendMessage;
 
-  // Send initial prompt if provided (for new sessions from homepage)
   useEffect(
     function handleInitialPrompt() {
       if (!initialPrompt || !sessionId) return;
@@ -241,15 +162,9 @@ export function useChatContainer({
 
       markInitialPromptSent(sessionId);
       const context = capturePageContext();
-      sendMessage(initialPrompt, true, context);
+      sendMessageRef.current(initialPrompt, true, context);
     },
-    [
-      initialPrompt,
-      sessionId,
-      initialMessages.length,
-      sendMessage,
-      capturePageContext,
-    ],
+    [initialPrompt, sessionId, initialMessages.length, capturePageContext],
   );
 
   async function sendMessageWithContext(

autogpt_platform/frontend/src/components/contextual/Chat/components/ChatInput/ChatInput.tsx

@@ -21,7 +21,7 @@ export function ChatInput({
   className,
 }: Props) {
   const inputId = "chat-input";
-  const { value, setValue, handleKeyDown, handleSend, hasMultipleLines } =
+  const { value, handleKeyDown, handleSubmit, handleChange, hasMultipleLines } =
     useChatInput({
       onSend,
       disabled: disabled || isStreaming,
@@ -29,15 +29,6 @@ export function ChatInput({
       inputId,
     });
 
-  function handleSubmit(e: React.FormEvent<HTMLFormElement>) {
-    e.preventDefault();
-    handleSend();
-  }
-
-  function handleChange(e: React.ChangeEvent<HTMLTextAreaElement>) {
-    setValue(e.target.value);
-  }
-
   return (
     <form onSubmit={handleSubmit} className={cn("relative flex-1", className)}>
       <div className="relative">

autogpt_platform/frontend/src/components/contextual/Chat/components/ChatInput/useChatInput.ts

@@ -1,4 +1,10 @@
-import { KeyboardEvent, useCallback, useEffect, useState } from "react";
+import {
+  ChangeEvent,
+  FormEvent,
+  KeyboardEvent,
+  useEffect,
+  useState,
+} from "react";
 
 interface UseChatInputArgs {
   onSend: (message: string) => void;
@@ -16,6 +22,23 @@ export function useChatInput({
   const [value, setValue] = useState("");
   const [hasMultipleLines, setHasMultipleLines] = useState(false);
 
+  useEffect(
+    function focusOnMount() {
+      const textarea = document.getElementById(inputId) as HTMLTextAreaElement;
+      if (textarea) textarea.focus();
+    },
+    [inputId],
+  );
+
+  useEffect(
+    function focusWhenEnabled() {
+      if (disabled) return;
+      const textarea = document.getElementById(inputId) as HTMLTextAreaElement;
+      if (textarea) textarea.focus();
+    },
+    [disabled, inputId],
+  );
+
   useEffect(() => {
     const textarea = document.getElementById(inputId) as HTMLTextAreaElement;
     const wrapper = document.getElementById(
@@ -77,7 +100,7 @@ export function useChatInput({
     }
   }, [value, maxRows, inputId]);
 
-  const handleSend = useCallback(() => {
+  const handleSend = () => {
     if (disabled || !value.trim()) return;
     onSend(value.trim());
     setValue("");
@@ -93,23 +116,31 @@ export function useChatInput({
       wrapper.style.height = "";
       wrapper.style.maxHeight = "";
     }
-  }, [value, onSend, disabled, inputId]);
+  };
 
-  const handleKeyDown = useCallback(
-    (event: KeyboardEvent<HTMLTextAreaElement>) => {
-      if (event.key === "Enter" && !event.shiftKey) {
-        event.preventDefault();
-        handleSend();
-      }
-    },
-    [handleSend],
-  );
+  function handleKeyDown(event: KeyboardEvent<HTMLTextAreaElement>) {
+    if (event.key === "Enter" && !event.shiftKey) {
+      event.preventDefault();
+      handleSend();
+    }
+  }
+
+  function handleSubmit(e: FormEvent<HTMLFormElement>) {
+    e.preventDefault();
+    handleSend();
+  }
+
+  function handleChange(e: ChangeEvent<HTMLTextAreaElement>) {
+    setValue(e.target.value);
+  }
 
   return {
     value,
     setValue,
     handleKeyDown,
     handleSend,
+    handleSubmit,
+    handleChange,
     hasMultipleLines,
   };
 }

autogpt_platform/frontend/src/components/contextual/Chat/components/MessageList/components/LastToolResponse/LastToolResponse.tsx

@@ -1,7 +1,5 @@
-import { AIChatBubble } from "../../../AIChatBubble/AIChatBubble";
 import type { ChatMessageData } from "../../../ChatMessage/useChatMessage";
-import { MarkdownContent } from "../../../MarkdownContent/MarkdownContent";
-import { formatToolResponse } from "../../../ToolResponseMessage/helpers";
+import { ToolResponseMessage } from "../../../ToolResponseMessage/ToolResponseMessage";
 import { shouldSkipAgentOutput } from "../../helpers";
 
 export interface LastToolResponseProps {
@@ -15,16 +13,15 @@ export function LastToolResponse({
 }: LastToolResponseProps) {
   if (message.type !== "tool_response") return null;
 
-  // Skip if this is an agent_output that should be rendered inside assistant message
   if (shouldSkipAgentOutput(message, prevMessage)) return null;
 
-  const formattedText = formatToolResponse(message.result, message.toolName);
-
   return (
     <div className="min-w-0 overflow-x-hidden hyphens-auto break-words px-4 py-2">
-      <AIChatBubble>
-        <MarkdownContent content={formattedText} />
-      </AIChatBubble>
+      <ToolResponseMessage
+        toolId={message.toolId}
+        toolName={message.toolName}
+        result={message.result}
+      />
     </div>
   );
 }

autogpt_platform/frontend/src/components/contextual/Chat/components/ToolResponseMessage/ToolResponseMessage.tsx

@@ -1,7 +1,14 @@
+import { Text } from "@/components/atoms/Text/Text";
+import { cn } from "@/lib/utils";
 import type { ToolResult } from "@/types/chat";
+import { WarningCircleIcon } from "@phosphor-icons/react";
 import { AIChatBubble } from "../AIChatBubble/AIChatBubble";
 import { MarkdownContent } from "../MarkdownContent/MarkdownContent";
-import { formatToolResponse } from "./helpers";
+import {
+  formatToolResponse,
+  getErrorMessage,
+  isErrorResponse,
+} from "./helpers";
 
 export interface ToolResponseMessageProps {
   toolId?: string;
@@ -18,6 +25,24 @@ export function ToolResponseMessage({
   success: _success,
   className,
 }: ToolResponseMessageProps) {
+  if (isErrorResponse(result)) {
+    const errorMessage = getErrorMessage(result);
+    return (
+      <AIChatBubble className={className}>
+        <div className="flex items-center gap-2">
+          <WarningCircleIcon
+            size={14}
+            weight="regular"
+            className="shrink-0 text-neutral-400"
+          />
+          <Text variant="small" className={cn("text-xs text-neutral-500")}>
+            {errorMessage}
+          </Text>
+        </div>
+      </AIChatBubble>
+    );
+  }
+
   const formattedText = formatToolResponse(result, toolName);
 
   return (

autogpt_platform/frontend/src/components/contextual/Chat/components/ToolResponseMessage/helpers.ts

@@ -1,3 +1,42 @@
+function stripInternalReasoning(content: string): string {
+  return content
+    .replace(/<internal_reasoning>[\s\S]*?<\/internal_reasoning>/gi, "")
+    .replace(/<thinking>[\s\S]*?<\/thinking>/gi, "")
+    .replace(/\n{3,}/g, "\n\n")
+    .trim();
+}
+
+export function isErrorResponse(result: unknown): boolean {
+  if (typeof result === "string") {
+    const lower = result.toLowerCase();
+    return (
+      lower.startsWith("error:") ||
+      lower.includes("not found") ||
+      lower.includes("does not exist") ||
+      lower.includes("failed to") ||
+      lower.includes("unable to")
+    );
+  }
+  if (typeof result === "object" && result !== null) {
+    const response = result as Record<string, unknown>;
+    return response.type === "error" || response.error !== undefined;
+  }
+  return false;
+}
+
+export function getErrorMessage(result: unknown): string {
+  if (typeof result === "string") {
+    return stripInternalReasoning(result.replace(/^error:\s*/i, ""));
+  }
+  if (typeof result === "object" && result !== null) {
+    const response = result as Record<string, unknown>;
+    if (response.error) return stripInternalReasoning(String(response.error));
+    if (response.message)
+      return stripInternalReasoning(String(response.message));
+  }
+  return "An error occurred";
+}
+
 function getToolCompletionPhrase(toolName: string): string {
   const toolCompletionPhrases: Record<string, string> = {
     add_understanding: "Updated your business information",
@@ -28,10 +67,10 @@ export function formatToolResponse(result: unknown, toolName: string): string {
         const parsed = JSON.parse(trimmed);
         return formatToolResponse(parsed, toolName);
       } catch {
-        return trimmed;
+        return stripInternalReasoning(trimmed);
       }
     }
-    return result;
+    return stripInternalReasoning(result);
   }
 
   if (typeof result !== "object" || result === null) {

autogpt_platform/frontend/src/components/contextual/Chat/stream-executor.ts

@@ -0,0 +1,142 @@
+import type {
+  ActiveStream,
+  StreamChunk,
+  VercelStreamChunk,
+} from "./chat-types";
+import {
+  INITIAL_RETRY_DELAY,
+  MAX_RETRIES,
+  normalizeStreamChunk,
+  parseSSELine,
+} from "./stream-utils";
+
+function notifySubscribers(stream: ActiveStream, chunk: StreamChunk) {
+  stream.chunks.push(chunk);
+  for (const callback of stream.onChunkCallbacks) {
+    try {
+      callback(chunk);
+    } catch (err) {
+      console.warn("[StreamExecutor] Subscriber callback error:", err);
+    }
+  }
+}
+
+export async function executeStream(
+  stream: ActiveStream,
+  message: string,
+  isUserMessage: boolean,
+  context?: { url: string; content: string },
+  retryCount: number = 0,
+): Promise<void> {
+  const { sessionId, abortController } = stream;
+
+  try {
+    const url = `/api/chat/sessions/${sessionId}/stream`;
+    const body = JSON.stringify({
+      message,
+      is_user_message: isUserMessage,
+      context: context || null,
+    });
+
+    const response = await fetch(url, {
+      method: "POST",
+      headers: {
+        "Content-Type": "application/json",
+        Accept: "text/event-stream",
+      },
+      body,
+      signal: abortController.signal,
+    });
+
+    if (!response.ok) {
+      const errorText = await response.text();
+      throw new Error(errorText || `HTTP ${response.status}`);
+    }
+
+    if (!response.body) {
+      throw new Error("Response body is null");
+    }
+
+    const reader = response.body.getReader();
+    const decoder = new TextDecoder();
+    let buffer = "";
+
+    while (true) {
+      const { done, value } = await reader.read();
+
+      if (done) {
+        notifySubscribers(stream, { type: "stream_end" });
+        stream.status = "completed";
+        return;
+      }
+
+      buffer += decoder.decode(value, { stream: true });
+      const lines = buffer.split("\n");
+      buffer = lines.pop() || "";
+
+      for (const line of lines) {
+        const data = parseSSELine(line);
+        if (data !== null) {
+          if (data === "[DONE]") {
+            notifySubscribers(stream, { type: "stream_end" });
+            stream.status = "completed";
+            return;
+          }
+
+          try {
+            const rawChunk = JSON.parse(data) as
+              | StreamChunk
+              | VercelStreamChunk;
+            const chunk = normalizeStreamChunk(rawChunk);
+            if (!chunk) continue;
+
+            notifySubscribers(stream, chunk);
+
+            if (chunk.type === "stream_end") {
+              stream.status = "completed";
+              return;
+            }
+
+            if (chunk.type === "error") {
+              stream.status = "error";
+              stream.error = new Error(
+                chunk.message || chunk.content || "Stream error",
+              );
+              return;
+            }
+          } catch (err) {
+            console.warn("[StreamExecutor] Failed to parse SSE chunk:", err);
+          }
+        }
+      }
+    }
+  } catch (err) {
+    if (err instanceof Error && err.name === "AbortError") {
+      notifySubscribers(stream, { type: "stream_end" });
+      stream.status = "completed";
+      return;
+    }
+
+    if (retryCount < MAX_RETRIES) {
+      const retryDelay = INITIAL_RETRY_DELAY * Math.pow(2, retryCount);
+      console.log(
+        `[StreamExecutor] Retrying in ${retryDelay}ms (attempt ${retryCount + 1}/${MAX_RETRIES})`,
+      );
+      await new Promise((resolve) => setTimeout(resolve, retryDelay));
+      return executeStream(
+        stream,
+        message,
+        isUserMessage,
+        context,
+        retryCount + 1,
+      );
+    }
+
+    stream.status = "error";
+    stream.error = err instanceof Error ? err : new Error("Stream failed");
+    notifySubscribers(stream, {
+      type: "error",
+      message: stream.error.message,
+    });
+  }
+}

autogpt_platform/frontend/src/components/contextual/Chat/stream-utils.ts

@@ -0,0 +1,84 @@
+import type { ToolArguments, ToolResult } from "@/types/chat";
+import type { StreamChunk, VercelStreamChunk } from "./chat-types";
+
+const LEGACY_STREAM_TYPES = new Set<StreamChunk["type"]>([
+  "text_chunk",
+  "text_ended",
+  "tool_call",
+  "tool_call_start",
+  "tool_response",
+  "login_needed",
+  "need_login",
+  "credentials_needed",
+  "error",
+  "usage",
+  "stream_end",
+]);
+
+export function isLegacyStreamChunk(
+  chunk: StreamChunk | VercelStreamChunk,
+): chunk is StreamChunk {
+  return LEGACY_STREAM_TYPES.has(chunk.type as StreamChunk["type"]);
+}
+
+export function normalizeStreamChunk(
+  chunk: StreamChunk | VercelStreamChunk,
+): StreamChunk | null {
+  if (isLegacyStreamChunk(chunk)) return chunk;
+
+  switch (chunk.type) {
+    case "text-delta":
+      return { type: "text_chunk", content: chunk.delta };
+    case "text-end":
+      return { type: "text_ended" };
+    case "tool-input-available":
+      return {
+        type: "tool_call_start",
+        tool_id: chunk.toolCallId,
+        tool_name: chunk.toolName,
+        arguments: chunk.input as ToolArguments,
+      };
+    case "tool-output-available":
+      return {
+        type: "tool_response",
+        tool_id: chunk.toolCallId,
+        tool_name: chunk.toolName,
+        result: chunk.output as ToolResult,
+        success: chunk.success ?? true,
+      };
+    case "usage":
+      return {
+        type: "usage",
+        promptTokens: chunk.promptTokens,
+        completionTokens: chunk.completionTokens,
+        totalTokens: chunk.totalTokens,
+      };
+    case "error":
+      return {
+        type: "error",
+        message: chunk.errorText,
+        code: chunk.code,
+        details: chunk.details,
+      };
+    case "finish":
+      return { type: "stream_end" };
+    case "start":
+    case "text-start":
+      return null;
+    case "tool-input-start":
+      return {
+        type: "tool_call_start",
+        tool_id: chunk.toolCallId,
+        tool_name: chunk.toolName,
+        arguments: {},
+      };
+  }
+}
+
+export const MAX_RETRIES = 3;
+export const INITIAL_RETRY_DELAY = 1000;
+
+export function parseSSELine(line: string): string | null {
+  if (line.startsWith("data: ")) return line.slice(6);
+  return null;
+}

autogpt_platform/frontend/src/components/contextual/Chat/useChat.ts

@@ -2,7 +2,6 @@
 
 import { useSupabase } from "@/lib/supabase/hooks/useSupabase";
 import { useEffect, useRef, useState } from "react";
-import { toast } from "sonner";
 import { useChatSession } from "./useChatSession";
 import { useChatStream } from "./useChatStream";
 
@@ -67,38 +66,16 @@ export function useChat({ urlSessionId }: UseChatArgs = {}) {
     ],
   );
 
-  useEffect(() => {
-    if (isLoading || isCreating) {
-      const timer = setTimeout(() => {
-        setShowLoader(true);
-      }, 300);
-      return () => clearTimeout(timer);
-    } else {
+  useEffect(
+    function showLoaderWithDelay() {
+      if (isLoading || isCreating) {
+        const timer = setTimeout(() => setShowLoader(true), 300);
+        return () => clearTimeout(timer);
+      }
       setShowLoader(false);
-    }
-  }, [isLoading, isCreating]);
-
-  useEffect(function monitorNetworkStatus() {
-    function handleOnline() {
-      toast.success("Connection restored", {
-        description: "You're back online",
-      });
-    }
-
-    function handleOffline() {
-      toast.error("You're offline", {
-        description: "Check your internet connection",
-      });
-    }
-
-    window.addEventListener("online", handleOnline);
-    window.addEventListener("offline", handleOffline);
-
-    return () => {
-      window.removeEventListener("online", handleOnline);
-      window.removeEventListener("offline", handleOffline);
-    };
-  }, []);
+    },
+    [isLoading, isCreating],
+  );
 
   function clearSession() {
     clearSessionBase();

autogpt_platform/frontend/src/components/contextual/Chat/useChatDrawer.ts

@@ -1,17 +0,0 @@
-"use client";
-
-import { create } from "zustand";
-
-interface ChatDrawerState {
-  isOpen: boolean;
-  open: () => void;
-  close: () => void;
-  toggle: () => void;
-}
-
-export const useChatDrawer = create<ChatDrawerState>((set) => ({
-  isOpen: false,
-  open: () => set({ isOpen: true }),
-  close: () => set({ isOpen: false }),
-  toggle: () => set((state) => ({ isOpen: !state.isOpen })),
-}));

autogpt_platform/frontend/src/components/contextual/Chat/useChatSession.ts

@@ -1,6 +1,7 @@
 import {
   getGetV2GetSessionQueryKey,
   getGetV2GetSessionQueryOptions,
+  getGetV2ListSessionsQueryKey,
   postV2CreateSession,
   useGetV2GetSession,
   usePatchV2SessionAssignUser,
@@ -101,6 +102,17 @@ export function useChatSession({
     }
   }, [createError, loadError]);
 
+  useEffect(
+    function refreshSessionsListOnLoad() {
+      if (sessionId && sessionData && !isLoadingSession) {
+        queryClient.invalidateQueries({
+          queryKey: getGetV2ListSessionsQueryKey(),
+        });
+      }
+    },
+    [sessionId, sessionData, isLoadingSession, queryClient],
+  );
+
   async function createSession() {
     try {
       setError(null);

autogpt_platform/frontend/src/components/contextual/Chat/useChatStream.ts

@@ -1,543 +1,110 @@
-import type { ToolArguments, ToolResult } from "@/types/chat";
-import { useCallback, useEffect, useRef, useState } from "react";
+"use client";
+
+import { useEffect, useRef, useState } from "react";
 import { toast } from "sonner";
+import { useChatStore } from "./chat-store";
+import type { StreamChunk } from "./chat-types";
 
-const MAX_RETRIES = 3;
-const INITIAL_RETRY_DELAY = 1000;
-
-export interface StreamChunk {
-  type:
-    | "text_chunk"
-    | "text_ended"
-    | "tool_call"
-    | "tool_call_start"
-    | "tool_response"
-    | "login_needed"
-    | "need_login"
-    | "credentials_needed"
-    | "error"
-    | "usage"
-    | "stream_end";
-  timestamp?: string;
-  content?: string;
-  message?: string;
-  code?: string;
-  details?: Record<string, unknown>;
-  tool_id?: string;
-  tool_name?: string;
-  arguments?: ToolArguments;
-  result?: ToolResult;
-  success?: boolean;
-  idx?: number;
-  session_id?: string;
-  agent_info?: {
-    graph_id: string;
-    name: string;
-    trigger_type: string;
-  };
-  provider?: string;
-  provider_name?: string;
-  credential_type?: string;
-  scopes?: string[];
-  title?: string;
-  [key: string]: unknown;
-}
-
-type VercelStreamChunk =
-  | { type: "start"; messageId: string }
-  | { type: "finish" }
-  | { type: "text-start"; id: string }
-  | { type: "text-delta"; id: string; delta: string }
-  | { type: "text-end"; id: string }
-  | { type: "tool-input-start"; toolCallId: string; toolName: string }
-  | {
-      type: "tool-input-available";
-      toolCallId: string;
-      toolName: string;
-      input: ToolArguments;
-    }
-  | {
-      type: "tool-output-available";
-      toolCallId: string;
-      toolName?: string;
-      output: ToolResult;
-      success?: boolean;
-    }
-  | {
-      type: "usage";
-      promptTokens: number;
-      completionTokens: number;
-      totalTokens: number;
-    }
-  | {
-      type: "error";
-      errorText: string;
-      code?: string;
-      details?: Record<string, unknown>;
-    };
-
-const LEGACY_STREAM_TYPES = new Set<StreamChunk["type"]>([
-  "text_chunk",
-  "text_ended",
-  "tool_call",
-  "tool_call_start",
-  "tool_response",
-  "login_needed",
-  "need_login",
-  "credentials_needed",
-  "error",
-  "usage",
-  "stream_end",
-]);
-
-function isLegacyStreamChunk(
-  chunk: StreamChunk | VercelStreamChunk,
-): chunk is StreamChunk {
-  return LEGACY_STREAM_TYPES.has(chunk.type as StreamChunk["type"]);
-}
-
-function normalizeStreamChunk(
-  chunk: StreamChunk | VercelStreamChunk,
-): StreamChunk | null {
-  if (isLegacyStreamChunk(chunk)) {
-    return chunk;
-  }
-  switch (chunk.type) {
-    case "text-delta":
-      return { type: "text_chunk", content: chunk.delta };
-    case "text-end":
-      return { type: "text_ended" };
-    case "tool-input-available":
-      return {
-        type: "tool_call_start",
-        tool_id: chunk.toolCallId,
-        tool_name: chunk.toolName,
-        arguments: chunk.input,
-      };
-    case "tool-output-available":
-      return {
-        type: "tool_response",
-        tool_id: chunk.toolCallId,
-        tool_name: chunk.toolName,
-        result: chunk.output,
-        success: chunk.success ?? true,
-      };
-    case "usage":
-      return {
-        type: "usage",
-        promptTokens: chunk.promptTokens,
-        completionTokens: chunk.completionTokens,
-        totalTokens: chunk.totalTokens,
-      };
-    case "error":
-      return {
-        type: "error",
-        message: chunk.errorText,
-        code: chunk.code,
-        details: chunk.details,
-      };
-    case "finish":
-      return { type: "stream_end" };
-    case "start":
-    case "text-start":
-      return null;
-    case "tool-input-start":
-      const toolInputStart = chunk as Extract<
-        VercelStreamChunk,
-        { type: "tool-input-start" }
-      >;
-      return {
-        type: "tool_call_start",
-        tool_id: toolInputStart.toolCallId,
-        tool_name: toolInputStart.toolName,
-        arguments: {},
-      };
-  }
-}
+export type { StreamChunk } from "./chat-types";
 
 export function useChatStream() {
   const [isStreaming, setIsStreaming] = useState(false);
   const [error, setError] = useState<Error | null>(null);
-  const retryCountRef = useRef<number>(0);
-  const retryTimeoutRef = useRef<NodeJS.Timeout | null>(null);
-  const abortControllerRef = useRef<AbortController | null>(null);
   const currentSessionIdRef = useRef<string | null>(null);
-  const requestStartTimeRef = useRef<number | null>(null);
-
-  const stopStreaming = useCallback(
-    (sessionId?: string, force: boolean = false) => {
-      console.log("[useChatStream] stopStreaming called", {
-        hasAbortController: !!abortControllerRef.current,
-        isAborted: abortControllerRef.current?.signal.aborted,
-        currentSessionId: currentSessionIdRef.current,
-        requestedSessionId: sessionId,
-        requestStartTime: requestStartTimeRef.current,
-        timeSinceStart: requestStartTimeRef.current
-          ? Date.now() - requestStartTimeRef.current
-          : null,
-        force,
-        stack: new Error().stack,
-      });
-
-      if (
-        sessionId &&
-        currentSessionIdRef.current &&
-        currentSessionIdRef.current !== sessionId
-      ) {
-        console.log(
-          "[useChatStream] Session changed, aborting previous stream",
-          {
-            oldSessionId: currentSessionIdRef.current,
-            newSessionId: sessionId,
-          },
-        );
-      }
-
-      const controller = abortControllerRef.current;
-      if (controller) {
-        const timeSinceStart = requestStartTimeRef.current
-          ? Date.now() - requestStartTimeRef.current
-          : null;
-
-        if (!force && timeSinceStart !== null && timeSinceStart < 100) {
-          console.log(
-            "[useChatStream] Request just started (<100ms), skipping abort to prevent race condition",
-            {
-              timeSinceStart,
-            },
-          );
-          return;
-        }
-
-        try {
-          const signal = controller.signal;
-
-          if (
-            signal &&
-            typeof signal.aborted === "boolean" &&
-            !signal.aborted
-          ) {
-            console.log("[useChatStream] Aborting stream");
-            controller.abort();
-          } else {
-            console.log(
-              "[useChatStream] Stream already aborted or signal invalid",
-            );
-          }
-        } catch (error) {
-          if (error instanceof Error && error.name === "AbortError") {
-            console.log(
-              "[useChatStream] AbortError caught (expected during cleanup)",
-            );
-          } else {
-            console.warn("[useChatStream] Error aborting stream:", error);
-          }
-        } finally {
-          abortControllerRef.current = null;
-          requestStartTimeRef.current = null;
-        }
-      }
-      if (retryTimeoutRef.current) {
-        clearTimeout(retryTimeoutRef.current);
-        retryTimeoutRef.current = null;
-      }
-      setIsStreaming(false);
-    },
-    [],
+  const onChunkCallbackRef = useRef<((chunk: StreamChunk) => void) | null>(
+    null,
   );
 
+  const stopStream = useChatStore((s) => s.stopStream);
+  const unregisterActiveSession = useChatStore(
+    (s) => s.unregisterActiveSession,
+  );
+  const isSessionActive = useChatStore((s) => s.isSessionActive);
+  const onStreamComplete = useChatStore((s) => s.onStreamComplete);
+  const getCompletedStream = useChatStore((s) => s.getCompletedStream);
+  const registerActiveSession = useChatStore((s) => s.registerActiveSession);
+  const startStream = useChatStore((s) => s.startStream);
+  const getStreamStatus = useChatStore((s) => s.getStreamStatus);
+
+  function stopStreaming(sessionId?: string) {
+    const targetSession = sessionId || currentSessionIdRef.current;
+    if (targetSession) {
+      stopStream(targetSession);
+      unregisterActiveSession(targetSession);
+    }
+    setIsStreaming(false);
+  }
+
   useEffect(() => {
-    console.log("[useChatStream] Component mounted");
-    return () => {
-      const sessionIdAtUnmount = currentSessionIdRef.current;
-      console.log(
-        "[useChatStream] Component unmounting, calling stopStreaming",
-        {
-          sessionIdAtUnmount,
-        },
-      );
-      stopStreaming(undefined, false);
+    return function cleanup() {
+      const sessionId = currentSessionIdRef.current;
+      if (sessionId && !isSessionActive(sessionId)) {
+        stopStream(sessionId);
+      }
       currentSessionIdRef.current = null;
+      onChunkCallbackRef.current = null;
     };
-  }, [stopStreaming]);
+  }, []);
 
-  const sendMessage = useCallback(
-    async (
-      sessionId: string,
-      message: string,
-      onChunk: (chunk: StreamChunk) => void,
-      isUserMessage: boolean = true,
-      context?: { url: string; content: string },
-      isRetry: boolean = false,
-    ) => {
-      console.log("[useChatStream] sendMessage called", {
-        sessionId,
-        message: message.substring(0, 50),
-        isUserMessage,
-        isRetry,
-        stack: new Error().stack,
-      });
+  useEffect(() => {
+    const unsubscribe = onStreamComplete(
+      function handleStreamComplete(completedSessionId) {
+        if (completedSessionId !== currentSessionIdRef.current) return;
 
-      const previousSessionId = currentSessionIdRef.current;
-      stopStreaming(sessionId, true);
-      currentSessionIdRef.current = sessionId;
-
-      const abortController = new AbortController();
-      abortControllerRef.current = abortController;
-      requestStartTimeRef.current = Date.now();
-      console.log("[useChatStream] Created new AbortController", {
-        sessionId,
-        previousSessionId,
-        requestStartTime: requestStartTimeRef.current,
-      });
-
-      if (abortController.signal.aborted) {
-        console.warn(
-          "[useChatStream] AbortController was aborted before request started",
-        );
-        requestStartTimeRef.current = null;
-        return Promise.reject(new Error("Request aborted"));
-      }
-
-      if (!isRetry) {
-        retryCountRef.current = 0;
-      }
-      setIsStreaming(true);
-      setError(null);
-
-      try {
-        const url = `/api/chat/sessions/${sessionId}/stream`;
-        const body = JSON.stringify({
-          message,
-          is_user_message: isUserMessage,
-          context: context || null,
-        });
-
-        const response = await fetch(url, {
-          method: "POST",
-          headers: {
-            "Content-Type": "application/json",
-            Accept: "text/event-stream",
-          },
-          body,
-          signal: abortController.signal,
-        });
-
-        console.info("[useChatStream] Stream response", {
-          sessionId,
-          status: response.status,
-          ok: response.ok,
-          contentType: response.headers.get("content-type"),
-        });
-
-        if (!response.ok) {
-          const errorText = await response.text();
-          console.warn("[useChatStream] Stream response error", {
-            sessionId,
-            status: response.status,
-            errorText,
-          });
-          throw new Error(errorText || `HTTP ${response.status}`);
-        }
-
-        if (!response.body) {
-          console.warn("[useChatStream] Response body is null", { sessionId });
-          throw new Error("Response body is null");
-        }
-
-        const reader = response.body.getReader();
-        const decoder = new TextDecoder();
-        let buffer = "";
-        let receivedChunkCount = 0;
-        let firstChunkAt: number | null = null;
-        let loggedLineCount = 0;
-
-        return new Promise<void>((resolve, reject) => {
-          let didDispatchStreamEnd = false;
-
-          function dispatchStreamEnd() {
-            if (didDispatchStreamEnd) return;
-            didDispatchStreamEnd = true;
-            onChunk({ type: "stream_end" });
-          }
-
-          const cleanup = () => {
-            reader.cancel().catch(() => {
-              // Ignore cancel errors
-            });
-          };
-
-          async function readStream() {
-            try {
-              while (true) {
-                const { done, value } = await reader.read();
-
-                if (done) {
-                  cleanup();
-                  console.info("[useChatStream] Stream closed", {
-                    sessionId,
-                    receivedChunkCount,
-                    timeSinceStart: requestStartTimeRef.current
-                      ? Date.now() - requestStartTimeRef.current
-                      : null,
-                  });
-                  dispatchStreamEnd();
-                  retryCountRef.current = 0;
-                  stopStreaming();
-                  resolve();
-                  return;
-                }
-
-                buffer += decoder.decode(value, { stream: true });
-                const lines = buffer.split("\n");
-                buffer = lines.pop() || "";
-
-                for (const line of lines) {
-                  if (line.startsWith("data: ")) {
-                    const data = line.slice(6);
-                    if (loggedLineCount < 3) {
-                      console.info("[useChatStream] Raw stream line", {
-                        sessionId,
-                        data:
-                          data.length > 300 ? `${data.slice(0, 300)}...` : data,
-                      });
-                      loggedLineCount += 1;
-                    }
-                    if (data === "[DONE]") {
-                      cleanup();
-                      console.info("[useChatStream] Stream done marker", {
-                        sessionId,
-                        receivedChunkCount,
-                        timeSinceStart: requestStartTimeRef.current
-                          ? Date.now() - requestStartTimeRef.current
-                          : null,
-                      });
-                      dispatchStreamEnd();
-                      retryCountRef.current = 0;
-                      stopStreaming();
-                      resolve();
-                      return;
-                    }
-
-                    try {
-                      const rawChunk = JSON.parse(data) as
-                        | StreamChunk
-                        | VercelStreamChunk;
-                      const chunk = normalizeStreamChunk(rawChunk);
-                      if (!chunk) {
-                        continue;
-                      }
-
-                      if (!firstChunkAt) {
-                        firstChunkAt = Date.now();
-                        console.info("[useChatStream] First stream chunk", {
-                          sessionId,
-                          chunkType: chunk.type,
-                          timeSinceStart: requestStartTimeRef.current
-                            ? firstChunkAt - requestStartTimeRef.current
-                            : null,
-                        });
-                      }
-                      receivedChunkCount += 1;
-
-                      // Call the chunk handler
-                      onChunk(chunk);
-
-                      // Handle stream lifecycle
-                      if (chunk.type === "stream_end") {
-                        didDispatchStreamEnd = true;
-                        cleanup();
-                        console.info("[useChatStream] Stream end chunk", {
-                          sessionId,
-                          receivedChunkCount,
-                          timeSinceStart: requestStartTimeRef.current
-                            ? Date.now() - requestStartTimeRef.current
-                            : null,
-                        });
-                        retryCountRef.current = 0;
-                        stopStreaming();
-                        resolve();
-                        return;
-                      } else if (chunk.type === "error") {
-                        cleanup();
-                        reject(
-                          new Error(
-                            chunk.message || chunk.content || "Stream error",
-                          ),
-                        );
-                        return;
-                      }
-                    } catch (err) {
-                      // Skip invalid JSON lines
-                      console.warn("Failed to parse SSE chunk:", err, data);
-                    }
-                  }
-                }
-              }
-            } catch (err) {
-              if (err instanceof Error && err.name === "AbortError") {
-                cleanup();
-                dispatchStreamEnd();
-                stopStreaming();
-                resolve();
-                return;
-              }
-
-              const streamError =
-                err instanceof Error ? err : new Error("Failed to read stream");
-
-              if (retryCountRef.current < MAX_RETRIES) {
-                retryCountRef.current += 1;
-                const retryDelay =
-                  INITIAL_RETRY_DELAY * Math.pow(2, retryCountRef.current - 1);
-
-                toast.info("Connection interrupted", {
-                  description: `Retrying in ${retryDelay / 1000} seconds...`,
-                });
-
-                retryTimeoutRef.current = setTimeout(() => {
-                  sendMessage(
-                    sessionId,
-                    message,
-                    onChunk,
-                    isUserMessage,
-                    context,
-                    true,
-                  ).catch((_err) => {
-                    // Retry failed
-                  });
-                }, retryDelay);
-              } else {
-                setError(streamError);
-                toast.error("Connection Failed", {
-                  description:
-                    "Unable to connect to chat service. Please try again.",
-                });
-                cleanup();
-                dispatchStreamEnd();
-                retryCountRef.current = 0;
-                stopStreaming();
-                reject(streamError);
-              }
-            }
-          }
-
-          readStream();
-        });
-      } catch (err) {
-        if (err instanceof Error && err.name === "AbortError") {
-          setIsStreaming(false);
-          return Promise.resolve();
-        }
-        const streamError =
-          err instanceof Error ? err : new Error("Failed to start stream");
-        setError(streamError);
         setIsStreaming(false);
-        throw streamError;
+        const completed = getCompletedStream(completedSessionId);
+        if (completed?.error) {
+          setError(completed.error);
+        }
+        unregisterActiveSession(completedSessionId);
+      },
+    );
+
+    return unsubscribe;
+  }, []);
+
+  async function sendMessage(
+    sessionId: string,
+    message: string,
+    onChunk: (chunk: StreamChunk) => void,
+    isUserMessage: boolean = true,
+    context?: { url: string; content: string },
+  ) {
+    const previousSessionId = currentSessionIdRef.current;
+    if (previousSessionId && previousSessionId !== sessionId) {
+      stopStreaming(previousSessionId);
+    }
+
+    currentSessionIdRef.current = sessionId;
+    onChunkCallbackRef.current = onChunk;
+    setIsStreaming(true);
+    setError(null);
+
+    registerActiveSession(sessionId);
+
+    try {
+      await startStream(sessionId, message, isUserMessage, context, onChunk);
+
+      const status = getStreamStatus(sessionId);
+      if (status === "error") {
+        const completed = getCompletedStream(sessionId);
+        if (completed?.error) {
+          setError(completed.error);
+          toast.error("Connection Failed", {
+            description: "Unable to connect to chat service. Please try again.",
+          });
+          throw completed.error;
+        }
       }
-    },
-    [stopStreaming],
-  );
+    } catch (err) {
+      const streamError =
+        err instanceof Error ? err : new Error("Failed to start stream");
+      setError(streamError);
+      throw streamError;
+    } finally {
+      setIsStreaming(false);
+    }
+  }
 
   return {
     isStreaming,

autogpt_platform/frontend/src/providers/posthog/posthog-provider.tsx

@@ -9,11 +9,12 @@ import { ReactNode, useEffect, useRef } from "react";
 
 export function PostHogProvider({ children }: { children: ReactNode }) {
   const isPostHogEnabled = environment.isPostHogEnabled();
+  const postHogCredentials = environment.getPostHogCredentials();
 
   useEffect(() => {
-    if (process.env.NEXT_PUBLIC_POSTHOG_KEY) {
-      posthog.init(process.env.NEXT_PUBLIC_POSTHOG_KEY, {
-        api_host: process.env.NEXT_PUBLIC_POSTHOG_HOST,
+    if (postHogCredentials.key) {
+      posthog.init(postHogCredentials.key, {
+        api_host: postHogCredentials.host,
         defaults: "2025-11-30",
         capture_pageview: false,
         capture_pageleave: true,

autogpt_platform/frontend/src/services/network-status/NetworkStatusMonitor.tsx

@@ -0,0 +1,8 @@
+"use client";
+
+import { useNetworkStatus } from "./useNetworkStatus";
+
+export function NetworkStatusMonitor() {
+  useNetworkStatus();
+  return null;
+}

autogpt_platform/frontend/src/services/network-status/useNetworkStatus.ts

@@ -0,0 +1,28 @@
+"use client";
+
+import { useEffect } from "react";
+import { toast } from "sonner";
+
+export function useNetworkStatus() {
+  useEffect(function monitorNetworkStatus() {
+    function handleOnline() {
+      toast.success("Connection restored", {
+        description: "You're back online",
+      });
+    }
+
+    function handleOffline() {
+      toast.error("You're offline", {
+        description: "Check your internet connection",
+      });
+    }
+
+    window.addEventListener("online", handleOnline);
+    window.addEventListener("offline", handleOffline);
+
+    return function cleanup() {
+      window.removeEventListener("online", handleOnline);
+      window.removeEventListener("offline", handleOffline);
+    };
+  }, []);
+}