## Summary
Fixes critical issue where `GET
/graphs/{graph_id}/executions/{graph_exec_id}` failed for marketplace
agents with "Graph not found" errors due to incorrect version access
checking.
## Root Cause
The endpoint was checking access to the **latest version** of a graph
instead of the **specific version used in the execution**. This broke
marketplace agents when:
1. User executes a marketplace agent (e.g., v3)
2. Graph owner later publishes a new version (e.g., v4)
3. User tries to view execution details
4. **BUG**: Code checked access to latest version (v4) instead of
execution version (v3)
5. If v4 wasn't published to marketplace → access denied → "Graph not
found"
## Original Problematic Code
```python
# routers/v1.py - get_graph_execution (WRONG ORDER)
graph = await graph_db.get_graph(graph_id=graph_id, user_id=user_id) # ❌ Uses LATEST version
if not graph:
raise HTTPException(404, f"Graph #{graph_id} not found")
result = await execution_db.get_graph_execution(...) # Gets execution data
```
## Solution
**Reordered operations** to check access against the **execution's
specific version**:
```python
# NEW CODE (CORRECT ORDER)
result = await execution_db.get_graph_execution(...) # ✅ Get execution FIRST
if not await graph_db.get_graph(
graph_id=result.graph_id,
version=result.graph_version, # ✅ Use execution's version, not latest!
user_id=user_id,
):
raise HTTPException(404, f"Graph #{graph_id} not found")
```
### Key Changes Made
1. **Fixed version access logic** (routers/v1.py:1075-1095):
- Reordered operations to get execution data first
- Check access using `result.graph_version` instead of latest version
- Applied same fix to external API routes
2. **Enhanced `get_graph()` marketplace fallback**
(data/graph.py:919-935):
- Added proper marketplace lookup when user doesn't own the graph
- Supports version-specific marketplace access checking
- Maintains security by only allowing approved, non-deleted listings
3. **Activity status generator fix**
(activity_status_generator.py:139-144):
- Use `skip_access_check=True` for internal system operations
4. **Missing block handling** (data/graph.py:94-103):
- Added `_UnknownBlockBase` placeholder for graceful handling of deleted
blocks
## Example Scenario Fixed
1. **User**: Installs marketplace agent "Blog Writer" v3
2. **Owner**: Later publishes v4 (not to marketplace yet)
3. **User**: Runs the agent (executes v3)
4. **Before**: Viewing execution details fails because code checked v4
access
5. **After**: ✅ Viewing execution details works because code checks v3
access
## Impact
- ✅ **Marketplace agents work correctly**: Users can view execution
details for any marketplace agent version they've used
- ✅ **Backward compatibility**: Existing owned graphs continue working
- ✅ **Security maintained**: Only allows access to versions user
legitimately executed
- ✅ **Version-aware access control**: Proper access checking for
specific versions, not just latest
## Testing
- [x] Marketplace agents: Execution details now accessible for all
executed versions
- [x] Owned graphs: Continue working as before
- [x] Version scenarios: Access control works correctly for specific
versions
- [x] Missing blocks: Graceful handling without errors
**Root issue resolved**: Version mismatch between execution version and
access check version that was breaking marketplace agent execution
viewing.
---------
Co-authored-by: Claude <noreply@anthropic.com>
- Resolves#11390
This unbreaks the last step of the Builder tutorial :)
### Changes 🏗️
- Give `isSaving` time to propagate before calling dependent callback
`saveAndRun`
### Checklist 📋
#### For code changes:
- [x] I have clearly listed my changes in the PR description
- [x] I have made a test plan
- [x] I have tested my changes according to the test plan:
- [x] Run through Builder tutorial; Run (with implicit save) should work
at once
## Changes 🏗️
Fixed the logout errors by removing duplicate redirects. `serverLogout`
was calling `redirect("/login")` (which throws `NEXT_REDIRECT`), and
then `useSupabaseStore` was also calling `router.refresh()`, causing
conflicts.
Updated `serverLogout` to return a result object instead of redirecting,
and moved the redirect to the client using `router.push("/login")` after
logout completes. This removes the `NEXT_REDIRECT` error and ensures a
single redirect.
<img width="800" height="706" alt="Screenshot 2025-11-18 at 16 14 54"
src="https://github.com/user-attachments/assets/38e0e55c-f48d-4b25-a07b-d4729e229c70"
/>
Also addressed 401 errors during logout. Hooks like `useCredits` were
still making API calls after logout, causing "Authorization header is
missing" errors. Added a check in `_makeClientRequest` to detect
logout-in-progress and suppress authentication errors during that
window. This prevents console noise and avoids unnecessary error
handling.
<img width="800" height="742" alt="Screenshot 2025-11-18 at 16 14 45"
src="https://github.com/user-attachments/assets/6fb2270a-97a0-4411-9e5a-9b4b52117af3"
/>
## Checklist 📋
### For code changes:
- [x] I have clearly listed my changes in the PR description
- [x] I have made a test plan
- [x] I have tested my changes according to the test plan:
- [x] Log out of your account
- [x] There are no errors showing up on the browser devtools
This refactor improves developer experience (DX) by creating a more
maintainable and extensible architecture.
The previous `CustomNode` implementation had several issues:
- Code was duplicated across different node types (StandardNodeBlock,
OutputBlock, etc.)
- Poor separation of concerns with all logic in a single component
- Limited flexibility for handling different block types
- Inconsistent handle display logic across different node types
<img width="2133" height="831" alt="Screenshot 2025-11-12 at 9 25 10 PM"
src="https://github.com/user-attachments/assets/02864bba-9ffe-4629-98ab-1c43fa644844"
/>
## Changes 🏗️
- **Refactored CustomNode structure**:
- Extracted reusable components:
[`NodeContainer`](file:///Users/abhi/Documents/AutoGPT/autogpt_platform/frontend/src/app/(platform)/build/components/FlowEditor/nodes/CustomNode/components/NodeContainer.tsx),
[`NodeHeader`](file:///Users/abhi/Documents/AutoGPT/autogpt_platform/frontend/src/app/(platform)/build/components/FlowEditor/nodes/CustomNode/components/NodeHeader.tsx),
[`NodeAdvancedToggle`](file:///Users/abhi/Documents/AutoGPT/autogpt_platform/frontend/src/app/(platform)/build/components/FlowEditor/nodes/CustomNode/components/NodeAdvancedToggle.tsx),
[`WebhookDisclaimer`](file:///Users/abhi/Documents/AutoGPT/autogpt_platform/frontend/src/app/(platform)/build/components/FlowEditor/nodes/CustomNode/components/WebhookDisclaimer.tsx)
- Removed `StandardNodeBlock.tsx` and consolidated logic into
[`CustomNode.tsx`](file:///Users/abhi/Documents/AutoGPT/autogpt_platform/frontend/src/app/(platform)/build/components/FlowEditor/nodes/CustomNode/CustomNode.tsx)
- Moved
[`StickyNoteBlock`](file:///Users/abhi/Documents/AutoGPT/autogpt_platform/frontend/src/app/(platform)/build/components/FlowEditor/nodes/CustomNode/components/StickyNoteBlock.tsx)
to components folder for better organization
- **Added BlockUIType-specific logic**:
- Implemented conditional handle display based on block type (INPUT,
WEBHOOK, WEBHOOK_MANUAL blocks don't show handles)
- Added special handling for AGENT blocks with dynamic input/output
schemas
- Added webhook-specific disclaimer component with library agent
integration
- Fixed OUTPUT block's name field to not show input handle
- **Enhanced FormCreator**:
- Added `showHandles` prop for granular control
- Added `className` prop for styling flexibility (used for webhook
opacity)
- **Improved nodeStore**:
- Added `getNodeBlockUIType` method for retrieving node UI types
- **UI/UX improvements**:
- Fixed duplicate gap classes in
[`BuilderActions`](file:///Users/abhi/Documents/AutoGPT/autogpt_platform/frontend/src/app/(platform)/build/components/BuilderActions/BuilderActions.tsx)
- Added proper styling for webhook blocks (disabled state with reduced
opacity)
- Improved field template spacing for specific block types
## Checklist 📋
### For code changes:
- [x] I have clearly listed my changes in the PR description
- [x] I have made a test plan
- [x] I have tested my changes according to the test plan:
- [x] Create and test a standard node block with input/output handles
- [x] Create and test INPUT block (verify no input handles)
- [x] Create and test OUTPUT block (verify name field has no handle)
- [x] Create and test WEBHOOK block (verify disclaimer appears and form
is disabled)
- [x] Create and test AGENT block with custom schemas
- [x] Create and test sticky note block
- [x] Verify advanced toggle works for all node types
- [x] Test node execution badges display correctly
- [x] Verify node selection highlighting works
Currently when an agent fails validation during a scheduled run, we
raise an error then try again, regardless of why.
This change removed the agent schedule and notifies the user
### Changes 🏗️
- add schedule_id to the GraphExecutionJobArgs
- add agent_name to the GraphExecutionJobArgs
- Delete schedule on GraphValidationError
- Notify the user with a message that include the agent name
### Checklist 📋
#### For code changes:
- [x] I have clearly listed my changes in the PR description
- [x] I have made a test plan
- [x] I have tested my changes according to the test plan:
<!-- Put your test plan here: -->
- [x] I have ensured the scheduler tests work with these changes
This PR removes turnstile from the platform.
#### For code changes:
- [x] I have clearly listed my changes in the PR description
- [x] I have made a test plan
- [x] I have tested my changes according to the test plan:
<!-- Put your test plan here: -->
- [x] Test to make sure that turnstile is gone, it will be.
- [x] Test logging in with out turnstile to make sure it still works
- [x] Test registering a new account with out turnstile and it works
## Summary
- Replaced the question mark icon with explicit "Give Feedback" text in
the feedback button
- Applied consistent styling to match the "Tutorial" button
- Removed QuestionMarkCircledIcon dependency from TallyPopup component
## Motivation
Users reported not knowing what the question mark icon was for, which
prevented them from discovering the feedback feature. Making the button
text-based and explicit removes this confusion.
## Changes
- Removed `QuestionMarkCircledIcon` import and icon element
- Changed button to display only "Give Feedback" text
- Added consistent styling (height, rounded corners, background color)
to match Tutorial button
- Button text can wrap to two lines if needed for better readability
#### For code changes:
- [x] I have clearly listed my changes in the PR description
- [x] I have made a test plan
- [x] I have tested my changes according to the test plan:
<!-- Put your test plan here: -->
- [x] Check the UI to see that the question mark on the tally button has
been replaced with "Give Feedback"
Before
<img width="618" height="198" alt="image"
src="https://github.com/user-attachments/assets/0d4803eb-9a05-4a43-aaff-cc43b6d0cda4"
/>
After
<img width="298" height="126" alt="image"
src="https://github.com/user-attachments/assets/c1e1c3b5-94b4-4ad9-87e9-a0feca1143e3"
/>
---------
Co-authored-by: Abhimanyu Yadav <122007096+Abhi1992002@users.noreply.github.com>
## Summary
Adds a non-blocking warning banner to Login and Sign Up pages that
alerts mobile users about potential limitations in the mobile
experience.
## Changes
- Created `MobileWarningBanner` component in `src/components/auth/`
- Integrated banner into Login page (`/login`)
- Integrated banner into Sign Up page (`/signup`)
- Banner displays only on mobile devices (viewports < 768px)
- Uses existing `useBreakpoint` hook for responsive detection
## Design Details
- **Position**: Appears below the login/signup card (after the bottom
"Sign up"/"Log in" links)
- **Style**: Amber-themed warning banner with DeviceMobile icon
- **Message**:
- Title: "Heads up: AutoGPT works best on desktop"
- Description: "Some features may be limited on mobile. For the best
experience, consider switching to a desktop."
- **Behavior**: Non-blocking, no user interaction required
<img width="342" height="81" alt="image"
src="https://github.com/user-attachments/assets/b6584299-b388-4d8d-b951-02bd95915566"
/>
#### For code changes:
- [x] I have clearly listed my changes in the PR description
- [x] I have made a test plan
- [x] I have tested my changes according to the test plan:
<!-- Put your test plan here: -->
- [x] Verified banner appears on mobile viewports (< 768px)
- [x] Verified banner is hidden on desktop viewports (≥ 768px)
- [x] Tested on Login page
- [x] Tested on Sign Up page
<img width="342" height="758" alt="image"
src="https://github.com/user-attachments/assets/077b3e0a-ab9c-41c7-83b7-7ee80a3396fd"
/>
<img width="342" height="759" alt="image"
src="https://github.com/user-attachments/assets/77a64b28-748b-4d97-bd7c-67c55e5e9f22"
/>
---------
Co-authored-by: Abhimanyu Yadav <122007096+Abhi1992002@users.noreply.github.com>
## Changes 🏗️
### Issue 1: login/signup redirect conflict
There are 2 hooks, both on the login and signup pages, that attempt to
call `router.push` once a user logs in or is created.
The main offender seems to be this hook:
```tsx
useEffect(() => {
if (user) router.push("/");
}, [user]);
```
Which is in place on both pages to prevent logged-in users from
accessing `/login` or `/signup`. What happens is when a user signs up or
logs in, if they need onboarding, there is a `router.push` down the line
to redirect them there, which conflicts with the one done in this hook.
**Solution**
I moved the logic from that hook to the `middleware.ts`, which is a
better place for it... It won't conflict anymore with onboarding
redirects done in those pages
### Issue 2: onboarding server redirects
Potential race condition: both the server component and the client
`<OnboardingProvider />` perform redirects. The server component
redirects happen first, but if onboarding state changes after mount, the
provider can redirect again, causing rapid mount/unmount cycles.
**Solution**
Make all onboarding redirects central in `/onboarding` which is now a
client component do in client redirects only and displaying a spinner
while it does so.
## Checklist 📋
#### For code changes:
- [x] I have clearly listed my changes in the PR description
- [x] I have made a test plan
- [x] I have tested my changes according to the test plan:
- [x] Tested locally login/logout/signup and trying to access `/login`
and `/signup` being logged in
Source maps aren't being uploaded to Sentry, so debugging errors in
production is really hard.
### Changes 🏗️
- Fix config so source maps are found and uploaded to Sentry
- Disable deleting source maps after upload (so they are available in
the browser)
### Checklist 📋
#### For code changes:
- [x] I have clearly listed my changes in the PR description
- [x] I have made a test plan
- [x] I have tested my changes according to the test plan:
- [x] Tested locally
<!-- Clearly explain the need for these changes: -->
This PR enhances the visual feedback in the flow editor by adding
animated "beads" that travel along edges during execution. This provides
users with clear, real-time visualization of data flow and execution
progress through the graph, making it easier to understand which
connections are active and track execution state.
https://github.com/user-attachments/assets/df4a4650-8192-403f-a200-15f6af95e384
### Changes 🏗️
<!-- Concisely describe all of the changes made in this pull request:
-->
- **Added new edge data types and structure:**
- Added `CustomEdgeData` type with `isStatic`, `beadUp`, `beadDown`, and
`beadData` properties
- Created `CustomEdge` type extending XYEdge with custom data
- **Implemented bead animation components:**
- Added `JSBeads.tsx` - JavaScript-based animation component with
real-time updates
- Added `SVGBeads.tsx` - SVG-based animation component (for future
consideration)
- Added helper functions for path calculations and bead positioning
- **Updated edge rendering:**
- Modified `CustomEdge` component to display beads during execution
- Added static edge styling with dashed lines (`stroke-dasharray: 6`)
- Improved visual hierarchy with different stroke styles for
selected/unselected states
- **Refactored edge management:**
- Converted `edgeStore` from using `Connection` type to `CustomEdge`
type
- Added `updateEdgeBeads` and `resetEdgeBeads` methods for bead state
management
- Updated `copyPasteStore` to work with new edge structure
- **Added support for static outputs:**
- Added `staticOutput` property to `CustomNodeData`
- Static edges show continuous bead animation while regular edges show
one-time animation
### Checklist 📋
#### For code changes:
- [x] I have clearly listed my changes in the PR description
- [x] I have made a test plan
- [x] I have tested my changes according to the test plan:
- [x] Create a flow with multiple blocks and verify beads animate along
edges during execution
- [x] Test that beads increment when execution starts (`beadUp`) and
decrement when completed (`beadDown`)
- [x] Verify static edges display with dashed lines and continuous
animation
- [x] Confirm copy/paste operations preserve edge data and bead states
- [x] Test edge animations performance with complex graphs (10+ nodes)
- [x] Verify bead animations complete properly before disappearing
- [x] Test that multiple beads can animate on the same edge for
concurrent executions
- [x] Verify edge selection/deletion still works with new visualization
- [x] Test that bead state resets properly when starting new executions
## Changes 🏗️
- Clear backend_id when pasting blocks to prevent duplicate ID errors
- Add copy/paste functionality to new FlowEditor
- Ensure pasted blocks use newly generated UUIDs when saving
Fixes issue where copying and pasting blocks would fail with 'Unique
constraint failed' error because the old backend_id was being reused
instead of the new node ID.
## Checklist 📋
### For code changes:
- [x] I have clearly listed my changes in the PR description
- [x] I have made a test plan
- [x] I have tested my changes according to the test plan:
- [x] Add any block to the builder, save and run the graph and wait for
it to finish, then copy and paste the first block and paste it, try to
use it and it should now work and not have any issues/errors
https://github.com/user-attachments/assets/c24f9a9a-8e4f-4988-8731-cddc34a0da13
---------
Co-authored-by: Reinier van der Leer <pwuts@agpt.co>
- Resolves#11305
### Changes 🏗️
Make `AIListGeneratorBlock` more reliable:
- Leverage `AIStructuredResponseGenerator`'s robust
prompt/retry/validate logic
- Use JSON format instead of Python list format
- Add `force_json_output` toggle
- Fix output instructions in prompt (only string values allowed)
### Checklist 📋
#### For code changes:
- [x] I have clearly listed my changes in the PR description
- [x] I have made a test plan
- [x] I have tested my changes according to the test plan:
- [x] Works without `force_json_output`
- [x] Works with `force_json_output`
- [x] Retry mechanism works as intended
Use WebSocket notifications from the backend to display confetti.
### Changes 🏗️
- Send WebSocket notifications to the browser when new onboarding steps
are completed
- Handle WebSocket notifications events in the Wallet and use them
instead of frontend-based logic to play confetti (fixes confetti
appearing on every refresh)
- Scroll to newly completed tasks when wallet opens just before confetti
plays
- Fix: make `Run again` button complete `RE_RUN_AGENT` task
### Checklist 📋
#### For code changes:
- [x] I have clearly listed my changes in the PR description
- [x] I have made a test plan
- [x] I have tested my changes according to the test plan:
- [x] Confetti are displayed when previously uncompleted tasks are
completed
- [x] Confetti do not appear on page refresh
- [x] Wallet scrolls on open before confetti is displayed
- [x] `Run again` button completes `RE_RUN_AGENT` task
This PR fixes a flaky test issue in the signup flow where Playwright's
strict mode was failing due to duplicate heading elements on the
marketplace page.
### Problem
The test was failing intermittently with the following error:
```
Error: strict mode violation: getByText('Bringing you AI agents designed by thinkers from around the world') resolved to 2 elements
```
This occurred because the marketplace page contains two identical `<h3>`
elements with the same text, causing Playwright's strict mode to throw
an error when trying to select a single element.
### Checklist 📋
#### For code changes:
- [x] I have clearly listed my changes in the PR description
- [x] I have made a test plan
- [x] I have tested my changes according to the test plan:
- [x] Run E2E tests locally multiple times to ensure no flakiness
- [x] Check CI pipeline runs successfully
Fixes two related bugs in the agent scheduling UI that caused confusion
for users setting up recurring schedules:
1. **"on day nan of every month" display bug**: When scheduling an agent
to repeat every N days (e.g., "every 2 days"), the schedule info panel
incorrectly displayed "on day nan of every month" instead of the correct
"Every N days at HH:MM" format.
2. **Confusing time picker for hourly intervals**: When setting up a
schedule with "every N hours", the UI displayed a time picker labeled
"at 9 o'clock" which was confusing because the time setting is ignored
for hourly intervals. Users were unclear about what this setting meant
or if it had any effect.
### Changes 🏗️
**Fixed `humanizeCronExpression` function**
(`autogpt_platform/frontend/src/lib/cron-expression-utils.ts`):
- Reordered cron expression parsing logic to handle day intervals
(`*/N`) before monthly checks
- Added `!dayOfMonth.startsWith("*/")` guard to monthly and yearly
checks to prevent misinterpreting day intervals as monthly day lists
- This ensures expressions like `0 9 */2 * *` (every 2 days at 9:00) are
correctly displayed as "Every 2 days at 09:00" instead of "on day nan of
every month"
**Updated `CronScheduler` component**
(`autogpt_platform/frontend/src/app/(platform)/library/agents/[id]/components/AgentRunsView/components/ScheduleAgentModal/components/CronScheduler/CronScheduler.tsx`):
- Hide `TimeAt` component for custom intervals with unit "hours" (time
is ignored for hourly intervals)
- Pass context-aware label to `TimeAt`: "Starting at" for custom day
intervals, "At" for other frequencies
- This clarifies that the time setting is the starting time for day
intervals and removes confusion for hourly intervals
**Enhanced `TimeAt` component**
(`autogpt_platform/frontend/src/app/(platform)/library/agents/[id]/components/AgentRunsView/components/ScheduleAgentModal/components/CronScheduler/TimeAt.tsx`):
- Added optional `label` prop (defaults to "At") to allow context-aware
labeling
- Component now displays "Starting at" when used with custom day
intervals for better clarity
### Checklist 📋
#### For code changes:
- [x] I have clearly listed my changes in the PR description
- [x] I have made a test plan
- [x] I have tested my changes according to the test plan:
<!-- Put your test plan here: -->
- [x] Schedule an agent with "Custom" frequency, "Every 2 days" interval
- verify it displays as "Every 2 days at HH:MM" in the schedule info
panel (not "on day nan of every month")
- [x] Schedule an agent with "Monthly" frequency - verify it displays
correctly (e.g., "On day 1, 15 of every month at HH:MM")
<img width="845" height="388" alt="image"
src="https://github.com/user-attachments/assets/02ed0b73-bf5e-48fd-a7b0-6f4d4687eb13"
/>
<img width="839" height="374" alt="image"
src="https://github.com/user-attachments/assets/be62eee2-3fdd-4b20-aecf-669c3c6c6fb2"
/>
- Resolves#11345
### Changes 🏗️
- Move tool use routing logic from frontend to backend: routing info was
being baked into graph links by the frontend, inconsistently, causing
issues
- Rework tool use routing to use target node ID instead of target block
name
- Add a bit of magic to `NodeOutputs` component to show tool node title
instead of ID
DX:
- Removed `build` from `.prettierignore` -> re-enable formatting for
builder components
### Checklist 📋
#### For code changes:
- [x] I have clearly listed my changes in the PR description
- [x] I have made a test plan
- [x] I have tested my changes according to the test plan:
- [x] Use SDM block in a graph; verify it works
- [x] Use SDM block with agent executor block as tool; verify it works
- Tests for `parse_execution_output` pass (checked by CI)
## Summary
This PR fixes an issue where multiple keyboard save event listeners were
being registered when the same save hook was used in multiple
components, causing the graph to be saved multiple times (3x) when using
Ctrl/Cmd+S.
## Changes
- **Created a centralized `useSaveGraph` hook** in
`/hooks/useSaveGraph.ts` that encapsulates all graph saving logic
- **Refactored `useNewSaveControl`** to use the new centralized hook
instead of duplicating save logic
- **Updated `useRunGraph` and `useScheduleGraph`** to use the
centralized `useSaveGraph` hook directly
- **Simplified the save control component** by removing redundant logic
and using cleaner naming conventions
## Problem
The previous implementation had the save logic duplicated in
`useNewSaveControl`, and when this hook was used in multiple places
(NewSaveControl component, RunGraph, ScheduleGraph), each instance would
register its own keyboard event listener for Ctrl/Cmd+S. This caused:
- Multiple save requests being sent simultaneously
- "Unique constraint failed on the fields: ('id', 'version')" errors
from the backend
- Poor performance due to unnecessary re-renders
## Solution
By centralizing the save logic in a dedicated `useSaveGraph` hook:
- Save logic is now in one place, making it easier to maintain
- Components can use the save functionality without registering
duplicate event listeners
- The keyboard shortcut listener is only registered once in the
`useNewSaveControl` hook
- Other components (RunGraph, ScheduleGraph) can call `saveGraph`
directly without side effects
## Testing
- [x] I have clearly listed my changes in the PR description
- [x] I have made a test plan
- [x] I have tested my changes according to the test plan:
- [x] Verified Ctrl/Cmd+S saves the graph only once
- [x] Tested save functionality from Save Control popup
- [x] Confirmed Run Graph and Schedule Graph still save before execution
- [x] Verified no duplicate save requests in network tab
- [x] Checked that save toast notifications appear correctly
We need a way to differentiate between serious errors that cause on call
alerts and block errors.
This PR address this need by ensuring all errors that occur during
execution of a block are of Subtype BlockError
### Changes 🏗️
- Introduced BlockErrors and its subtypes
- Updated current errors that are emitted by blocks to use BlockError
- Update executor manager, to errors emitted when running a block that
are not of type BlockError to BlockUnknownError
### Checklist 📋
#### For code changes:
- [x] I have clearly listed my changes in the PR description
- [x] I have made a test plan
- [x] I have tested my changes according to the test plan:
<!-- Put your test plan here: -->
- [x] checked tests still work
- [x] Ensured block error message is readable and useful
In this PR, I’ve added drag-and-drop functionality to the new builder
using built-in HTML drag-and-drop.
https://github.com/user-attachments/assets/b27c281e-6216-4131-9a89-e10b0dd56a8f
### Changes
- Added ReactFlowProvider to manage flow state in BuilderPage and Flow
components.
- Implemented drag-and-drop support for blocks in the NewControlPanel,
allowing users to drag blocks from the menu and drop them onto the
canvas.
- Enhanced the Block component to handle drag events and provide visual
feedback during dragging.
- Updated useFlow hook to include onDragOver and onDrop handlers for
managing block placement.
- Adjusted nodeStore to accept position parameters for added blocks,
improving placement accuracy.
### Checklist 📋
#### For code changes:
- [x] I have clearly listed my changes in the PR description
- [x] I have made a test plan
- [x] I have tested my changes according to the test plan:
- [x] I’ve tried dragging and dropping multiple blocks, and it works
perfectly as shown in the video.
- depends one https://github.com/Significant-Gravitas/AutoGPT/pull/11339
In this PR, I’ve added a dropdown menu to the custom node. This allows
you to delete a node, copy a node, and if the node is an agent node, you
can also navigate to that specific agent.
<img width="633" height="403" alt="Screenshot 2025-11-08 at 7 24 38 PM"
src="https://github.com/user-attachments/assets/89dd2906-95f5-40a5-82d1-de05075e4f30"
/>
###Changes
- Added context menu to custom nodes with copy, delete, and open agent
options
- Added performance optimization with memo for custom edge
### Checklist 📋
#### For code changes:
- [x] I have clearly listed my changes in the PR description
- [x] I have made a test plan
- [x] I have tested my changes according to the test plan:
- [x] All three buttons are working perfectly.
- [x] The “Go to graph” option is only visible in the sub-graph node.
In the new builder, I’ve added a copy-paste functionality using the
keyboard.
https://github.com/user-attachments/assets/3106ae86-3f47-4807-a598-9c0b166eaae9
### Changes 🏗️
- Added useCopyPasteKeyboard hook for handling keyboard shortcuts
- Created new copyPasteStore for state management
- Implemented performance optimizations (memo on CustomEdge)
- Updated nodeStore and edgeStore to support the functionality
### Checklist 📋
#### For code changes:
- [x] I have clearly listed my changes in the PR description
- [x] I have made a test plan
- [x] I have tested my changes according to the test plan:
- [x] The copy-paste functionality is working correctly as you can see
in the video.
## Summary
Fix admin impersonation not working for graph execution requests that
are server-side rendered.
## Problem
- Build page uses SSR, so API calls go through _makeServerRequest
instead of _makeClientRequest
- Server-side requests cannot access sessionStorage where impersonation
ID is stored
- Graph execution requests were missing X-Act-As-User-Id header
## Simple Solution
1. **Store impersonation in cookie** (useAdminImpersonation.ts):
- Set/clear cookie alongside sessionStorage for server access
2. **Read cookie on server** (_makeServerRequest in client.ts):
- Check for impersonation cookie using Next.js cookies() API
- Create fake Request with X-Act-As-User-Id header
- Pass to existing makeAuthenticatedRequest flow
## Changes Made
- useAdminImpersonation.ts: 2 lines to set/clear cookie
- client.ts: 1 method to read cookie and create header
- No changes to existing proxy/header/helpers logic
## Result
- ✅ Graph execution requests now include impersonation header
- ✅ Works for both client-side and server-side rendered requests
- ✅ Minimal changes, leverages existing header forwarding logic
- ✅ Backward compatible with all existing functionality
🤖 Generated with [Claude Code](https://claude.ai/code)
Co-Authored-By: Claude <noreply@anthropic.com>
---------
Co-authored-by: Claude <noreply@anthropic.com>
### Changes 🏗️
This change uses [Zustand](https://github.com/pmndrs/zustand) (a
lightweight state management library) to centralize authentication state
across the app. Previously, each component mounting `useSupabase()`
would create its own local state, causing duplicate API calls and
inconsistent user data. Now, user state is cached globally with Zustand
- when multiple components need auth data, they share the same cached
state instead of each fetching separately. This reduces server load and
improves app responsiveness.
**File structure:**
```
src/lib/supabase/hooks/
├── useSupabase.ts # React hook interface (modified)
├── useSupabaseStore.ts # Zustand state management (new)
└── helpers.ts # Pure business logic (new)
```
**What was extracted to helpers:**
- `ensureSupabaseClient()` - Singleton client initialization
- `fetchUser()` - User fetching with error handling
- `validateSession()` - Session validation logic
- `refreshSession()` - Session refresh logic
- `handleStorageEvent()` - Cross-tab logout handling
### Checklist 📋
#### For code changes:
- [x] I have clearly listed my changes in the PR description
- [x] I have made a test plan
- [x] I have tested my changes according to the test plan:
- [x] Verified no TypeScript errors in modified files
- [x] Tested login flow works correctly
- [x] Tested logout flow works correctly
- [x] Verified session validation on tab focus/visibility
- [x] Tested cross-tab logout synchronization
- [x] Confirmed WebSocket disconnection on logout
This PR adds a comprehensive execution analytics admin endpoint that
generates AI-powered activity summaries and correctness scores for graph
executions, with proper feature flag bypass for admin use.
### Changes 🏗️
**Backend Changes:**
- Added admin endpoint: `/api/executions/admin/execution_analytics`
- Implemented feature flag bypass with `skip_feature_flag=True`
parameter for admin operations
- Fixed async database client usage (`get_db_async_client`) to resolve
async/await errors
- Added batch processing with configurable size limits to handle large
datasets
- Comprehensive error handling and logging for troubleshooting
- Renamed entire feature from "Activity Backfill" to "Execution
Analytics" for clarity
**Frontend Changes:**
- Created clean admin UI for execution analytics generation at
`/admin/execution-analytics`
- Built form with graph ID input, model selection dropdown, and optional
filters
- Implemented results table with status badges and detailed execution
information
- Added CSV export functionality for analytics results
- Integrated with generated TypeScript API client for proper
authentication
- Added proper error handling with toast notifications and loading
states
**Database & API:**
- Fixed critical async/await issue by switching from sync to async
database client
- Updated router configuration and endpoint naming for consistency
- Generated proper TypeScript types and API client integration
- Applied feature flag filtering at API level while bypassing for admin
operations
### Checklist 📋
#### For code changes:
- [x] I have clearly listed my changes in the PR description
- [x] I have made a test plan
- [x] I have tested my changes according to the test plan:
**Test Plan:**
- [x] Admin can access execution analytics page at
`/admin/execution-analytics`
- [x] Form validation works correctly (requires graph ID, validates
inputs)
- [x] API endpoint `/api/executions/admin/execution_analytics` responds
correctly
- [x] Authentication works properly through generated API client
- [x] Analytics generation works with different LLM models (gpt-4o-mini,
gpt-4o, etc.)
- [x] Results display correctly with appropriate status badges
(success/failed/skipped)
- [x] CSV export functionality downloads correct data
- [x] Error handling displays appropriate toast messages
- [x] Feature flag bypass works for admin users (generates analytics
regardless of user flags)
- [x] Batch processing handles multiple executions correctly
- [x] Loading states show proper feedback during processing
#### For configuration changes:
- [x] `.env.default` is updated or already compatible with my changes
- [x] `docker-compose.yml` is updated or already compatible with my
changes
- [x] No configuration changes required for this feature
**Related to:** PR #11325 (base correctness score functionality)
🤖 Generated with [Claude Code](https://claude.ai/code)
---------
Co-authored-by: Claude <noreply@anthropic.com>
Co-authored-by: claude[bot] <41898282+claude[bot]@users.noreply.github.com>
Co-authored-by: Zamil Majdy <majdyz@users.noreply.github.com>
This enables real time notifications from backend to browser via
WebSocket using Redis bus for moving notifications from REST process to
WebSocket process.
This is needed for (follow-up) backend-completion of onboarding tasks
with instant notifications.
### Changes 🏗️
- Add new `AsyncRedisNotificationEventBus` to enable publishing
notifications to the Redis event bus
- Consume notifications in `ws_api.py` similarly to execution events and
send them via WebSocket
- Store WebSocket user connections in `ConnectionManager`
- Add relevant tests and types
### Checklist 📋
#### For code changes:
- [x] I have clearly listed my changes in the PR description
- [x] I have made a test plan
- [x] I have tested my changes according to the test plan:
- [x] Notifications are sent to the frontend
## Changes 🏗️
Allow dynamic URLs in the CORS config, to match them via regex. This
helps because currently we have Front-end preview deployments which are
isolated ( _nice they don't pollute or overrride other domains_ ) like:
```
https://autogpt-git-{branch_name}-{commit}-significant-gravitas.vercel.app
```
The Front-end builds and works there, but as soon as you login, any API
requests to endpoints that need auth will fail due to CORS, given our
current CORS config does not support dynamically generated domains.
### Changes
After these changes we can specify dynamic domains to be allowed under
CORS. I also made `localhost` disabled if the API is in production for
safety...
### Before
```yml
cors:
allowOrigin: "https://dev-builder.agpt.co" # could only specify full URL strings, not dyamic ones
```
### After
```yml
cors:
allowOrigins:
- "https://dev-builder.agpt.co"
- "regex:https://autogpt-git-[a-z0-9-]+\\.vercel\\.app" # dynamic domains supported via regex
```
### Files
- add `build_cors_params` utility to parse literal/regex origins and
block localhost in production (`backend/server/utils/cors.py`)
- apply the helper in both `AgentServer` and `WebsocketServer` so CORS
logic and validations remain consistent
- add reusable `override_config` testing helper and update existing
WebSocket tests to cover the shared CORS behavior
- introduce targeted unit tests for the new CORS helper
(`backend/server/utils/cors_test.py`)
## Checklist 📋
#### For code changes:
- [x] I have clearly listed my changes in the PR description
- [x] I have made a test plan
- [x] I have tested my changes according to the test plan:
- [x] We will know once we made the origin config changes on infra and
test with this...
## Changes 🏗️
Make sure we can login on preview deployments generated by Vercel to
test Front-end changes. As of now, the Cloudflare CAPTCHA verification
fails, we don't need to have it active there.
### Minor improvements
<img width="1599" height="755" alt="Screenshot 2025-11-06 at 16 18 10"
src="https://github.com/user-attachments/assets/0a3fb1f3-2d4d-49fe-885f-10f141dc0ce4"
/>
Prevent the following build error:
```
15:58:01.507
at j (.next/server/app/(no-navbar)/onboarding/reset/page.js:1:5125)
15:58:01.507
at <unknown> (.next/server/chunks/5826.js:2:14221)
15:58:01.507
at b.handleCallbackErrors (.next/server/chunks/5826.js:43:43068)
15:58:01.507
at <unknown> (.next/server/chunks/5826.js:2:14194) {
15:58:01.507
description: "Route /onboarding/reset couldn't be rendered statically because it used `cookies`. See more info here: https://nextjs.org/docs/messages/dynamic-server-error",
15:58:01.507
digest: 'DYNAMIC_SERVER_USAGE'
15:58:01.507
}
```
by making the reset onboarding route a client one. I made a new component, `<LoadingSpinner />`, and that page will show it while onboarding it's being reset.
## Checklist 📋
### For code changes:
- [x] I have clearly listed my changes in the PR description
- [x] I have made a test plan
- [x] I have tested my changes according to the test plan:
- [x] You can login/signup on the app and use it in the preview URL generated by Vercel
These models have become deprecated
- deepseek-r1-distill-llama-70b
- gemma2-9b-it
- llama3-70b-8192
- llama3-8b-8192
- google/gemini-flash-1.5
I have removed them and setup a migration, the migration is to convert
all the old versions of the model to new versions, the model changes
will happen like so
- llama3-70b-8192 → llama-3.3-70b-versatile
- llama3-8b-8192 → llama-3.1-8b-instant
- google/gemini-flash-1.5 → google/gemini-2.5-flash
- deepseek-r1-distill-llama-70b → gpt-5-chat-latest
- gemma2-9b-it → gpt-5-chat-latest
### Changes 🏗️
<!-- Concisely describe all of the changes made in this pull request:
-->
### Checklist 📋
#### For code changes:
- [x] I have clearly listed my changes in the PR description
- [x] I have made a test plan
- [x] I have tested my changes according to the test plan:
<!-- Put your test plan here: -->
- [x] Check to see if old models where removed
- [x] Check to see if migration worked and converted old models to new
one in graph
### Changes 🏗️
- Replaces `isSupersetOf` and `difference` Set operations with
backwards-compatible implementations using `Array.from` and
`every`/`filter` methods.
- This ensures compatibility with older JavaScript environments that may
not fully support modern Set operations.
Fixes
[BUILDER-451](https://sentry.io/organizations/significant-gravitas/issues/6952591149/).
The issue was that: ES2024 Set methods `isSupersetOf` and `difference`
are unsupported in iOS Safari 16.7, causing a TypeError during component
render.
This fix was generated by Seer in Sentry, triggered automatically. 👁️
Run ID: 2032240
Not quite right? [Click here to continue debugging with
Seer.](https://sentry.io/organizations/significant-gravitas/issues/6952591149/?seerDrawer=true)
### Checklist 📋
#### For code changes:
- [ ] I have clearly listed my changes in the PR description
- [ ] I have made a test plan
- [ ] I have tested my changes according to the test plan:
<!-- Put your test plan here: -->
- [ ] Test on iOS Safari 16.7 to ensure no TypeError occurs during
component render.
- [ ] Verify that the replaced `isSupersetOf` and `difference`
implementations function correctly in other supported browsers.
<details>
<summary>Example test plan</summary>
- [ ] Create from scratch and execute an agent with at least 3 blocks
- [ ] Import an agent from file upload, and confirm it executes
correctly
- [ ] Upload agent to marketplace
- [ ] Import an agent from marketplace and confirm it executes correctly
- [ ] Edit an agent from monitor, and confirm it executes correctly
</details>
#### For configuration changes:
- [ ] `.env.default` is updated or already compatible with my changes
- [ ] `docker-compose.yml` is updated or already compatible with my
changes
- [ ] I have included a list of my configuration changes in the PR
description (under **Changes**)
<details>
<summary>Examples of configuration changes</summary>
- Changing ports
- Adding new services that need to communicate with each other
- Secrets or environment variable changes
- New or infrastructure changes such as databases
</details>
Co-authored-by: seer-by-sentry[bot] <157164994+seer-by-sentry[bot]@users.noreply.github.com>
Implements a cookie consent banner and settings modal for GDPR
compliance, allowing users to manage preferences for analytics and
monitoring cookies. Integrates consent checks with Sentry, Vercel
Analytics, and Google Analytics, ensuring tracking is only enabled with
user permission. Refactors dialog components for improved layout and
adds consent management utilities and hooks.
#### For code changes:
- [x] Banner appears at bottom of page on first visit with rounded
corners and proper spacing (40px margins)
- [x] Banner shows three buttons: "Reject All", "Settings", and "Accept
All"
- [x] Clicking "Accept All" hides banner and enables
analytics/monitoring
- [x] Clicking "Reject All" hides banner and keeps analytics/monitoring
disabled
- [x] Banner does not reappear after consent is given (check
localStorage: `autogpt_cookie_consent`)
**Cookie Settings Modal:**
- [x] Clicking "Settings" button opens the Cookie Settings modal
- [x] Modal displays three categories: Essential Cookies (always
active), Analytics & Performance (toggle), Error Monitoring & Session
Replay (toggle)
- [x] Clicking "Save Preferences" saves custom settings and closes modal
- [x] Clicking "Accept All" enables all cookies and closes modal
- [x] Clicking "Reject All" disables optional cookies and closes modal
- [x] Modal can be closed with X button or clicking outside
**Consent Persistence:**
- [x] Refresh page after giving consent - banner should not reappear
- [x] Clear localStorage and refresh - banner should reappear
- [x] Consent choices persist across browser sessions
<img width="1123" height="126" alt="image"
src="https://github.com/user-attachments/assets/7425efab-b5cc-4449-802d-0e12bd65053b"
/>
<img width="1124" height="372" alt="image"
src="https://github.com/user-attachments/assets/2f28919a-97e8-44f5-9021-70d3836bb996"
/>
<!-- Clearly explain the need for these changes: -->
Fixes
[BUILDER-48G](https://sentry.io/organizations/significant-gravitas/issues/6960009111/).
The issue was that: Asynchronous API update scheduled via
`setTimeout(0)` in `OnboardingProvider` creates a race condition,
causing React-DOM's portal cleanup (`removeChild`) to fail during
concurrent component unmounting.
### Changes 🏗️
<!-- Concisely describe all of the changes made in this pull request:
-->
- Prevents state updates and API calls after the `OnboardingProvider`
component has been unmounted.
- Introduces a `isMounted` ref to track the component's mount status.
- Uses a `pendingUpdatesRef` to manage and cancel pending API update
promises on unmount, preventing memory leaks and errors.
- Ensures that API update errors are only logged if the component is
still mounted.
This fix was generated by Seer in Sentry, triggered by Craig Swift. 👁️
Run ID: 2058387
Not quite right? [Click here to continue debugging with
Seer.](https://sentry.io/organizations/significant-gravitas/issues/6960009111/?seerDrawer=true)
### Checklist 📋
#### For code changes:
- [x] I have clearly listed my changes in the PR description
- [x] I have made a test plan
- [x] I have tested my changes according to the test plan:
- [x] Onboarding works and does not throw errors when unmounted
---------
Co-authored-by: seer-by-sentry[bot] <157164994+seer-by-sentry[bot]@users.noreply.github.com>
Co-authored-by: Lluis Agusti <hi@llu.lu>
Co-authored-by: Ubbe <hi@ubbe.dev>
- Resolves#11314
### Changes 🏗️
- Change "Download agent" CTA button to action link at bottom of summary
agent info
- Move agent ratings above CTA buttons to prevent it from jumping on
page load
- Update vertical spacings to more closely match designs
### Checklist 📋
#### For code changes:
- [x] I have clearly listed my changes in the PR description
- [x] I have made a test plan
- [x] I have tested my changes according to the test plan:
- [x] Designer approves of new look
- [x] Tests pass
### Changes 🏗️
Fixes
[BUILDER-4HJ](https://sentry.io/organizations/significant-gravitas/issues/6979388537/).
The issue was that: Server-side rendering failed to retrieve the
Supabase access token, causing authenticated API calls to omit the
Authorization header.
- Ensures that the agent version is fetched only when
`creator_agent.active_version_id` exists and the status code is 200.
- Enables the `prefetchGetV2GetAgentByStoreIdQuery` query when
`creator_agent.active_version_id` exists.
This fix was generated by Seer in Sentry, triggered by Craig Swift. 👁️
Run ID: 2234004
Not quite right? [Click here to continue debugging with
Seer.](https://sentry.io/organizations/significant-gravitas/issues/6979388537/?seerDrawer=true)
### Checklist 📋
#### For code changes:
- [x] I have clearly listed my changes in the PR description
- [x] I have made a test plan
- [x] I have tested my changes according to the test plan:
- [x] Loading marketplace works...
Co-authored-by: seer-by-sentry[bot] <157164994+seer-by-sentry[bot]@users.noreply.github.com>
Co-authored-by: Ubbe <hi@ubbe.dev>
## Changes 🏗️
<img width="800" height="800" alt="Screenshot 2025-11-04 at 23 05 22"
src="https://github.com/user-attachments/assets/ecb3f442-8f1b-4a80-a6c9-0c4b6d5e0427"
/>
New `<Button variant="link" />` for when you need to render an HTML
`<button>` but with our link styles.
## Checklist 📋
### For code changes:
- [x] I have clearly listed my changes in the PR description
- [x] I have made a test plan
- [x] I have tested my changes according to the test plan:
- [x] Run Storybook locally
- [x] Looks good
## Summary
Add AI-generated correctness score field to execution activity status
generation to provide quantitative assessment of how well executions
achieved their intended purpose.
New page:
<img width="1000" height="229" alt="image"
src="https://github.com/user-attachments/assets/5cb907cf-5bc7-4b96-8128-8eecccde9960"
/>
Old page:
<img width="1000" alt="image"
src="https://github.com/user-attachments/assets/ece0dfab-1e50-4121-9985-d585f7fcd4d2"
/>
## What Changed
- Added `correctness_score` field (float 0.0-1.0) to
`GraphExecutionStats` model
- **REFACTORED**: Removed duplicate `llm_utils.py` and reused existing
`AIStructuredResponseGeneratorBlock` logic
- Updated activity status generator to use structured responses instead
of plain text
- Modified prompts to include correctness assessment with 5-tier scoring
system:
- 0.0-0.2: Failure
- 0.2-0.4: Poor
- 0.4-0.6: Partial Success
- 0.6-0.8: Mostly Successful
- 0.8-1.0: Success
- Updated manager.py to extract and set both activity_status and
correctness_score
- Fixed tests to work with existing structured response interface
## Technical Details
- **Code Reuse**: Eliminated duplication by using existing
`AIStructuredResponseGeneratorBlock` instead of creating new LLM
utilities
- Added JSON validation with retry logic for malformed responses
- Maintained backward compatibility for existing activity status
functionality
- Score is clamped to valid 0.0-1.0 range and validated
- All type errors resolved and linting passes
## Test Plan
- [x] All existing tests pass with refactored structure
- [x] Structured LLM call functionality tested with success and error
cases
- [x] Activity status generation tested with various execution scenarios
- [x] Integration tests verify both fields are properly set in execution
stats
- [x] No code duplication - reuses existing block logic
🤖 Generated with [Claude Code](https://claude.ai/code)
---------
Co-authored-by: Claude <noreply@anthropic.com>
Co-authored-by: claude[bot] <41898282+claude[bot]@users.noreply.github.com>
Co-authored-by: Zamil Majdy <majdyz@users.noreply.github.com>
BREAKING CHANGE: Removed deprecated use_auto_prompt field from Input
schema. Existing workflows using this field will need to be updated to
use the type field set to "auto" instead.
## Summary of Changes 📝
This PR comprehensively updates all Exa search blocks to match the
latest Exa API specification and adds significant new functionality
through the Websets API integration.
### Core API Updates 🔄
- **Migration to Exa SDK**: Replaced manual API calls with the official
`exa_py` AsyncExa SDK across all blocks for better reliability and
maintainability
- **Removed deprecated fields**: Eliminated
`use_auto_prompt`/`useAutoprompt` field (breaking change)
- **Fixed incomplete field definitions**: Corrected `user_location`
field definition
- **Added new input fields**: Added `moderation` and `context` fields
for enhanced content filtering
### Enhanced Content Settings 🛠️
- **Text field improvements**: Support both boolean and advanced object
configurations
- **New content options**:
- Added `livecrawl` settings (never, fallback, always, preferred)
- Added `subpages` support for deeper content retrieval
- Added `extras` settings for links and images
- Added `context` settings for additional contextual information
- **Updated settings**: Enhanced `highlight` and `summary`
configurations with new query and schema options
### Comprehensive Cost Tracking 💰
- Added detailed cost tracking models:
- `CostDollars` for monetary costs
- `CostCredits` for API credit tracking
- `CostDuration` for time-based costs
- New output fields: `request_id`, `resolved_search_type`,
`cost_dollars`
- Improved response handling to conditionally yield fields based on
availability
### New Websets API Integration 🚀
Added eight new specialized blocks for Exa's Websets API:
- **`websets.py`**: Core webset management (create, get, list, delete)
- **`websets_search.py`**: Search operations within websets
- **`websets_items.py`**: Individual item management (add, get, update,
delete)
- **`websets_enrichment.py`**: Data enrichment operations
- **`websets_import_export.py`**: Bulk import/export functionality
- **`websets_monitor.py`**: Monitor and track webset changes
- **`websets_polling.py`**: Poll for updates and changes
### New Special-Purpose Blocks 🎯
- **`code_context.py`**: Code search capabilities for finding relevant
code snippets from open source repositories, documentation, and Stack
Overflow
- **`research.py`**: Asynchronous research capabilities that explore the
web, gather sources, synthesize findings, and return structured results
with citations
### Code Organization Improvements 📁
- **Removed legacy code**: Deleted `model.py` file containing deprecated
API models
- **Centralized helpers**: Consolidated shared models and utilities in
`helpers.py`
- **Improved modularity**: Each webset operation is now in its own
dedicated file
### Other Changes 🔧
- Updated `.gitignore` for better development workflow
- Updated `CLAUDE.md` with project-specific instructions
- Updated documentation in `docs/content/platform/new_blocks.md` with
error handling, data models, and file input guidelines
- Improved webhook block implementations with SDK integration
### Files Changed 📂
- **Modified (11 files)**:
- `.gitignore`
- `autogpt_platform/CLAUDE.md`
- `autogpt_platform/backend/backend/blocks/exa/answers.py`
- `autogpt_platform/backend/backend/blocks/exa/contents.py`
- `autogpt_platform/backend/backend/blocks/exa/helpers.py`
- `autogpt_platform/backend/backend/blocks/exa/search.py`
- `autogpt_platform/backend/backend/blocks/exa/similar.py`
- `autogpt_platform/backend/backend/blocks/exa/webhook_blocks.py`
- `autogpt_platform/backend/backend/blocks/exa/websets.py`
- `docs/content/platform/new_blocks.md`
- **Added (8 files)**:
- `autogpt_platform/backend/backend/blocks/exa/code_context.py`
- `autogpt_platform/backend/backend/blocks/exa/research.py`
- `autogpt_platform/backend/backend/blocks/exa/websets_enrichment.py`
- `autogpt_platform/backend/backend/blocks/exa/websets_import_export.py`
- `autogpt_platform/backend/backend/blocks/exa/websets_items.py`
- `autogpt_platform/backend/backend/blocks/exa/websets_monitor.py`
- `autogpt_platform/backend/backend/blocks/exa/websets_polling.py`
- `autogpt_platform/backend/backend/blocks/exa/websets_search.py`
- **Deleted (1 file)**:
- `autogpt_platform/backend/backend/blocks/exa/model.py`
### Migration Guide 🚦
For users with existing workflows using the deprecated `use_auto_prompt`
field:
1. Remove the `use_auto_prompt` field from your input configuration
2. Set the `type` field to `ExaSearchTypes.AUTO` (or "auto" in JSON) to
achieve the same behavior
3. Review any custom content settings as the structure has been enhanced
### Testing Recommendations ✅
- Test existing workflows to ensure they handle the breaking change
- Verify cost tracking fields are properly returned
- Test new content settings options (livecrawl, subpages, extras,
context)
- Validate websets functionality if using the new Websets API blocks
🤖 Generated with [Claude Code](https://claude.com/claude-code)
### Checklist 📋
#### For code changes:
- [x] I have clearly listed my changes in the PR description
- [x] I have made a test plan
- [x] I have tested my changes according to the test plan:
- [x] made + ran a test agent for the blocks and flows between them
[Exa
Tests_v44.json](https://github.com/user-attachments/files/23226143/Exa.Tests_v44.json)
<!-- CURSOR_SUMMARY -->
---
> [!NOTE]
> Migrates Exa blocks to AsyncExa SDK, adds comprehensive
Websets/research/code-context blocks, updates existing
search/content/answers/similar, deletes legacy models, adjusts
tests/docs; breaking: remove `use_auto_prompt` in favor of
`type="auto"`.
>
> - **Backend — Exa integration (SDK migration & BREAKING)**:
> - Replace manual HTTP calls with `exa_py.AsyncExa` across `search`,
`similar`, `contents`, `answers`, and webhooks; richer outputs
(citations, context, costs, resolved search type).
> - BREAKING: remove `Input.use_auto_prompt`; use `type = "auto"`.
> - Centralize models/utilities in `exa/helpers.py` (content settings,
cost models, result mappers).
> - **New Blocks**:
> - **Websets**: management (`websets.py`), searches, items,
enrichments, imports/exports, monitors, polling (new files under
`exa/websets_*`).
> - **Research**: async research task create/get/wait/list
(`exa/research.py`).
> - **Code Context**: code snippet/context retrieval
(`exa/code_context.py`).
> - **Removals**:
> - Delete deprecated `exa/model.py`.
> - **Docs & DX**:
> - Update `docs/new_blocks.md` (error handling, models, file input) and
`CLAUDE.md`; ignore backend logs in `.gitignore`.
> - **Frontend Tests**:
> - Split/extend “e” block tests and improve block add robustness in
Playwright (`build.spec.ts`, `build.page.ts`).
>
> <sup>Written by [Cursor
Bugbot](https://cursor.com/dashboard?tab=bugbot) for commit
6e5e572322. This will update automatically
on new commits. Configure
[here](https://cursor.com/dashboard?tab=bugbot).</sup>
<!-- /CURSOR_SUMMARY -->
<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit
* **New Features**
* Added multiple Exa research and webset management blocks for task
creation, monitoring, and completion tracking.
* Introduced new search capabilities including code context retrieval,
content search, and enhanced filtering options.
* Added webset enrichment, import/export, and item management
functionality.
* Expanded search with location-based and category filters.
* **Documentation**
* Updated guidance on error handling, data models, and file input
handling.
* **Refactor**
* Modernized backend API integration with improved response structure
and error reporting.
* Simplified configuration options for search operations.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->
---------
Co-authored-by: Claude <noreply@anthropic.com>
- Resolves#11316
- Durable fix to replace #11318
### Changes 🏗️
- Expand graph execution permissions check
- Don't require library membership for execution as sub-graph
### Checklist 📋
#### For code changes:
- [x] I have clearly listed my changes in the PR description
- [x] I have made a test plan
- [x] I have tested my changes according to the test plan:
- [x] Can run sub-agent with non-latest graph version
- [x] Can run sub-agent that is available in Marketplace but not added
to Library
