## Context
The REST pipeline re-fetches/sets the metadata version later in
RestApiBaseHandler#getObjectMetadata by reading from DB and seeding the
cache if it’s missing. That’s the place that actually needs it and
already handles the “undefined” case.
This also mirror the graphql path that was updated 3 weeks ago
## Summary
- Add a context usage indicator to the AI chat interface inspired by
Vercel's AI SDK Context component
- Display token consumption, context window utilization percentage, and
estimated cost in credits
- Show a circular progress ring with percentage, revealing detailed
breakdown on hover
## Changes
### Backend
- Stream usage metadata (tokens, model config) via `messageMetadata`
callback in `agent-chat-streaming.service.ts`
- Return model config from `chat-execution.service.ts`
- Add usage and model types to `ExtendedUIMessage` metadata
### Frontend
- New `ContextUsageProgressRing` component - circular SVG progress
indicator
- New `AIChatContextUsageButton` component with hover card showing:
- Progress bar with used/total tokens
- Input/output token counts with credit costs
- Total credits consumed
- Track cumulative usage in Recoil state (`agentChatUsageState`)
- Reset usage when creating new chat thread
- Integrate button into `AIChatTab`
## Test plan
- [ ] Open AI chat and send a message
- [ ] Verify the context usage button appears with percentage
- [ ] Hover over the button to see detailed breakdown
- [ ] Verify credits are calculated correctly
- [ ] Create a new chat thread and verify usage resets to 0
## Summary
- Implements real tools for the dashboard-building skill to create and
manage dashboards through the AI chat interface
- Adds 6 new dashboard tools: `create_complete_dashboard`,
`list_dashboards`, `get_dashboard`, `add_dashboard_widget`,
`update_dashboard_widget`, `delete_dashboard_widget`
- Improves widget configuration robustness with typed Zod schemas and
discriminated unions for graph types
## Key Changes
**New Dashboard Tools:**
- `create_complete_dashboard` - Creates a dashboard with layout, tab,
and widgets in a single call
- `list_dashboards` - Lists all dashboards in the workspace
- `get_dashboard` - Gets full dashboard details including tabs and
widget configurations
- `add_dashboard_widget` - Adds a widget to an existing dashboard tab
- `update_dashboard_widget` - Updates widget properties or configuration
- `delete_dashboard_widget` - Removes a widget from a dashboard
**Widget Configuration Improvements:**
- Typed Zod schemas for each chart type (AGGREGATE, BAR, LINE, PIE)
- Discriminated union validation based on `graphType`
- Widget-level error handling for partial success when creating
dashboards
- Clear documentation about required `objectMetadataId` and field UUIDs
**Skill Documentation Updates:**
- Updated `dashboard-building.skill.ts` with critical guidance about
looking up field metadata first
- Added workflow instructions: use `list_object_metadata_items` before
creating GRAPH widgets
- Practical grid layout recommendations
## Test plan
- [ ] Create a new dashboard via AI chat
- [ ] Verify widgets display data correctly when proper field IDs are
provided
- [ ] Test adding/updating/deleting widgets on existing dashboards
- [ ] Verify error messages are helpful when configuration is incorrect
## Summary
- Replace the agent search mechanism with a new skills-based system
- Add a `skills` module with predefined skill definitions that the AI
can load on demand
- Remove specialized agents (workflow-builder, data-manipulator,
dashboard-builder, metadata-builder, researcher), keeping only the
helper agent
- Add `recordReferences` to workflow creation tool for chip linking in
the UI
## Changes
### New Skills Module
- `skill-definitions.ts` - Contains 5 skill definitions with detailed
instructions
- `skills.service.ts` - Service to get skills by name
- `load-skill.tool.ts` - Tool for AI to load skills explicitly
### Removed
- `agent-search.tool.ts` - Replaced by skill loading
- Specialized agent definitions (converted to skills)
### Updated
- Chat execution now shows skill catalog in system prompt
- Workflow creation returns `recordReferences` for UI linking
## Test plan
- [ ] Verify AI can load skills using `load_skill` tool
- [ ] Verify skill content is returned correctly
- [ ] Verify workflow creation shows clickable chip in chat
- [ ] Verify helper agent still works
## Summary
- Add latest AI models from OpenAI (GPT-4.1, o3, o4-mini), Anthropic
(Claude 4.5 Opus/Sonnet/Haiku), and xAI (Grok 4.1)
- Mark deprecated models (GPT-4o, GPT-4o-mini, GPT-4-turbo, Claude Opus
4, Claude Sonnet 4) with a `deprecated` flag
- Split AI models into separate files per provider for better
maintainability
- Support comma-separated default model lists for automatic fallback
across providers (works out of the box for self-hosters regardless of
which provider they configure)
- Filter deprecated models from dropdown selection while keeping them
functional for existing agents
## Changes
### New Models Added
| Provider | Models |
|----------|--------|
| OpenAI | gpt-4.1, gpt-4.1-mini, o3, o4-mini |
| Anthropic | claude-opus-4-5, claude-sonnet-4-5, claude-haiku-4-5 |
| xAI | grok-4-1-fast-reasoning |
### Deprecated Models
- gpt-4o, gpt-4o-mini, gpt-4-turbo (OpenAI)
- claude-opus-4-20250514, claude-sonnet-4-20250514 (Anthropic)
### Config Changes
Default model configs now support comma-separated fallback lists:
-
`DEFAULT_AI_SPEED_MODEL_ID=gpt-4.1-mini,claude-haiku-4-5-20251001,grok-3-mini`
-
`DEFAULT_AI_PERFORMANCE_MODEL_ID=gpt-4.1,claude-sonnet-4-5-20250929,grok-4`
## Test plan
- [x] Unit tests pass
- [x] Typecheck passes
- [x] Lint passes
- [ ] Verify deprecated models don't appear in model dropdowns
- [ ] Verify agents with deprecated models still work correctly
- [ ] Verify default model fallback works when only one provider is
configured
## Summary
Adds a new **VIEW** tool category for the AI chat, enabling it to work
with views:
- **get-views**: List views in the workspace, optionally filtered by
object metadata ID
- **get-view-query-parameters**: Convert a view's filters and sorts into
GraphQL query parameters that can be passed to existing `find_*` data
tools
- **create-view**, **update-view**, **delete-view**: CRUD operations for
view management
### Key design decisions
1. **No pagination duplication**: Instead of creating a
`find-records-from-view` tool that would duplicate pagination logic,
`get-view-query-parameters` returns filter/sort parameters that the AI
can pass to existing record-fetching tools.
2. **Permission model**:
- Read tools (get-views, get-view-query-parameters) are available to all
users
- Write tools require the `VIEW` permission
- UNLISTED views can only be modified by their creator
3. **Leverages existing utilities**: Uses
`computeRecordGqlOperationFilter` from `twenty-shared` for filter
conversion.
### Files changed
- Added `ViewToolProvider`, `ViewToolsFactory`, and
`ViewQueryParamsService`
- Added `VIEW` to `ToolCategory` enum and tool registry
- Updated `chat-execution.service.ts` to include view tools in the
catalog and pass viewId in browsing context
- Extracted shared `formatValidationErrors` utility to reduce
duplication
## Test plan
- [x] Unit tests for `ViewToolsFactory`
- [x] Unit tests for `ViewQueryParamsService`
- [x] Lint and typecheck pass
# TimelineActivity migration to morph
- Creates `timelineActivities2` relations on Company, Dashboard, Note,
Opportunity, Person, Task, Workflow, WorkflowRun, and WorkflowVersion
entities with proper metadata and cascade delete behavior.
It was required to create standard fields as well since the
mapObjectMetadataByUniqueIdentifier needs it. otherwise the fields won't
be considered
- Feature Flag `IS_TIMELINE_ACTIVITY_MIGRATED` necessary to have the two
states in parallel. It is used as a stamp once the migration has been
run
- Migration is done using the coreDataSource. Why ?
even though is unsafe to use, the first implementation of the migration
took forever on each workspace. See [this
commit](https://github.com/twentyhq/twenty/pull/15652/commits/477011e8d7d4c580f79ba7ec4a8fb002a3ec86b2)
The plan for this complex migration is as follows :
Note: we will need to rename fields in the release 1.12 (there is no
easy way to do all this in one release)
## Summary
- Add `BrowsingContext` type to automatically pass what the user is
currently viewing (recordPage or listView) to the AI chat
- Simplify context architecture: remove toggleable context UI, make it
automatic and invisible to the user
- Fix tool loading: add `unionOf` handling in
`getDatabaseToolsForObject` and fix regex ordering so `find_one_*` tools
are properly registered
- Use plural names for find tools (`find_people` vs `find_one_person`)
for better semantics
- Clean up unused components and states
## Changes
### Frontend
- New `BrowsingContext` type and `useGetBrowsingContext` hook to gather
context from Recoil state
- Simplified `useAgentChat` to use the new browsing context
- Removed toggleable context UI components
(`AgentChatContextRecordPreview`, `SendMessageWithRecordsContextButton`,
etc.)
- Removed `isAgentChatCurrentContextActiveState`
### Backend
- New `BrowsingContextType` for recordPage and listView contexts
- Updated `ChatExecutionService` to build context from browsing context
- Fixed `tool-registry.service.ts`:
- Added `unionOf` handling in permission config
- Fixed regex ordering (`find_one` before `find`) so tools load
correctly
- Use plural names for search tools (`find_people` instead of
`find_person`)
## Test plan
- [x] Typecheck passes
- [x] Lint passes
- [ ] Test AI chat on record page - should show context in system prompt
- [ ] Test AI chat on list view - should show view name and filters
- [ ] Test `find_one_*` tools now load correctly
- [ ] Test `find_*` tools use plural naming
The current repository implementation require the caller to pass the
permissionConfig which is not the case in messaging and calendar custom
resolver. We are bypassing the permission as fetching the role in the
caller will likely not be needed anymore in the new datasource / orm
implementation
## Context
Following https://github.com/twentyhq/twenty/pull/16399
Now using the new global orm manager everywhere and returning a
GlobalDatasource/WorkspaceDatasource based on a feature flag.
This means we now need to wrap all our ORM calls within
executeInWorkspaceContext callback (at least for now) so the global
datasource can dynamically hydrate its context via the new store (the
global datasource does not store anything related to workspaces as it is
now a unique singleton). If feature flag is off it still uses local data
stored in the workspace datasource.
## Context
SyncMetadata for role update was broken when those roles had permission
flags in them.
This PR https://github.com/twentyhq/twenty/issues/16365 introduced a
change in all existing standard roles which probably triggered the
update of faulty roles for the first time. Down the line, some
properties from flat were not omit before querying the DB (in this case
permissionFlagIds does not exist as a column in role table)
The fix is definitely not great, this also probably broke the update of
permissionFlags for existing STANDARD roles but it's not subject to
change (It actually never changed. Again, what has been updated was not
the permission flag but a new boolean introduced 2 days ago which forced
the "UPDATE" path in the sync which was probably never triggered before,
or at least not for roles with permission flags which are "quite" new).
Also we are about to remove this whole code in favor of the new
migration v2
# Introduction
Early returning in `validateBuildAndRun` as the runner even if passed an
empty workspace migration actions array be invalidating dependency
related flat entity maps.
When we will refactor the runner to consume the generic flat entity maps
tools such as
`addFlatEntityToFlatEntityAndRelatedEntityMapsThroughMutationOrThrow` we
will be able to remove the dependency cache invalidation
## Summary
This PR significantly simplifies the AI chat architecture by removing
complex routing/planning mechanisms and introduces clickable record
links in AI responses.
## Changes
### AI Chat Architecture Simplification
- **Removed** the entire `ai-chat-router` module (~850 lines) including:
- Strategy decider service
- Plan generator service
- Complex routing logic
- **Removed** agent execution planning services (~700 lines):
- `agent-execution.service.ts`
- `agent-plan-executor.service.ts`
- `agent-tool-generator.service.ts`
- **Added** centralized `ToolRegistryService` for tool management:
- Builds searchable tool index (database, action, workflow tools)
- Provides tool lookup by name
- Supports agent search for loading expertise
- **Added** `ChatExecutionService` as simple replacement:
- Includes full tool catalog in system prompt
- Pre-loads common tools (find/create/update for company, person,
opportunity, task, note)
- Uses `load_tools` mechanism for dynamic tool activation
- Enables native web search by default
### Record References in AI Responses
- Added `recordReferences` field to tool outputs for create, find, and
update operations
- Implemented `[[record:objectName:recordId:displayName]]` syntax for AI
to reference records
- Created `RecordLink` component that renders clickable chips with
object icons
- Integrated record link parsing into the markdown renderer
- Users can now click directly on created/found records in AI responses
### Workflow Agent Fixes
- Fixed cache invalidation issue when creating agents in workflows
- Added default prompt for workflow-created agents to prevent validation
errors
- Relaxed agent validation to only check properties being updated (not
all required properties)
### Code Quality Improvements
- Extracted `getRecordDisplayName` utility that mirrors frontend's
`getLabelIdentifierFieldValue` logic
- Uses object metadata to determine the correct label identifier field
- Handles `FULL_NAME` composite type for person/workspaceMember objects
- Shared across create, find, and update record services
## Net Impact
- **~1,200 lines deleted** (complex routing/planning code)
- **~500 lines added** (simpler tool registry + record links)
- Significantly reduced code complexity
- Better tool discovery through full catalog in system prompt
- Improved UX with clickable record references
## Testing
- Typecheck passes
- Lint passes
- Manual testing of AI chat with record creation and linking
# Introduction
In this PR we've introduced the type safe generation tooling for
standard view, view fields, view filters and view groups.
Also generated the ids before such as what was done for object and
fields, it's still temporary as will be replaced by universalIdentifier
mapping afterwards
## Task views
In this PR only implemented the `task` views and view groups, filters
and fields in order to battle test all the generation tools
Will create a dedicated PR that will implement all the remaining views.
- The issue is that the listener is not triggered if they are no change
to the workspaceMember (which happens when name is already filled
through GoogleSSO)
Fixes https://github.com/twentyhq/twenty/issues/16440
Changes:
- as we store date in redis as serialized, let's make all flatEntity
dates as string. This requires changing FlatEntity types and making sure
that entity are converted to flatEntity and flatEntity to dtos
# Introduction
Related https://github.com/twentyhq/core-team-issues/issues/1995
In this PR we're introducing the standard index tools for their
declaration
Also adding a common typing in order to have a simily consistency across
standard builders
Will implement remaining standard index builder in an other PR
Fixes https://github.com/twentyhq/twenty/issues/15925
- update field metadata update logic
- uniformize the way index are named
- command to migrate v1-named unique index
- add integration testing
---------
Co-authored-by: prastoin <paul@twenty.com>
Using deepEqual, average over 3 calls:
- version trigger 1.53ms
- version steps 38.3ms
- run state 372.8 ms
Using stringified hash, average over 3 calls:
- version trigger 0.07ms
- version steps 0.74 ms
- run state 0.521ms
Short fastDeepEqual
- version trigger 0.028ms
- version steps 0.197ms
- run state 0.214ms
Lib fastDeepEqual
- version trigger 0.038ms
- version steps 0.187ms
- run state 0.230ms
## Context
Deprecating TwentyORMManager in favor of TwentyORMGlobalManager
(temporarily, as this will simplify the ultimate goal to later replace
all usages with the new TwentyORMGlobalManagerV2 which will have a
similar signature)
This means this PR had to refactor a bit of code to pass down the
workspaceId when not available directly as it is now a requirement,
meaning we also deprecated scopedWorkspaceContextFactory to have a less
obscure way to fetch the workspaceId and have something more
declarative.
Step 3 will be to update TwentyORMGlobalManager to use a featureFlag
toggling and use the new GlobalWorkspaceOrmManager internally using the
new cache service
Step 4 will be to remove the feature flag and pg_pool patch
In this PR:
- add more logs to troubleshoot issues in production
- use messageChannelSyncStatus in messageSync service instead of making
a manual update
Abdullah.