## Problem
In production, an AI-chat assistant response sometimes freezes
mid-stream (partial text, looks hung), then "picks up again on its own"
later without the user resending and without a known worker restart.
Root cause: the **agent-chat SSE subscription has no keepalive and no
silent-death detection**.
- Delivery is fire-and-forget Redis pub/sub
(`SubscriptionService.publishToAgentChat`) and the resolver returns the
**raw** iterator — unlike `EventStreamResolver`, which heartbeats every
30s via `wrapAsyncIteratorWithLifecycle`.
- During a quiet model/tool gap the connection sends no bytes, so a
proxy/LB/NAT can silently drop it mid-stream. `graphql-sse` neither
surfaces an error nor resumes with `Last-Event-ID`, and **nothing
re-pulls the existing Redis chunk catch-up on reconnect** (it only runs
on thread (re)mount / `message-persisted` refetch).
- So the live view freezes; recovery only happens when the terminal
`message-persisted` fires a full refetch from the DB — the observed
"self-recovery".
This is the **same silent-SSE-death class fixed for the DB event stream
in #21061**, which was never applied to the agent-chat path. The symptom
also matches #21096 (worker logs the job finishing, client never
updates, reload shows the message).
It is **not** queue prioritization, and it is **not** addressed by
#22193 (which only stabilizes the assistant message id and removes
end-of-stream flicker).
A secondary, independent self-recovery path also existed: BullMQ
stalled-job re-run (default 30s `lockDuration`, no idempotency guard)
re-streaming the whole turn → duplicate assistant messages / double
billing.
## Changes
### Commit 1 — keepalive + silent-death recovery (ports the #21061
pattern to agent chat)
- **Shared:** new `keepalive` variant on `AgentChatSubscriptionEvent`.
- **Server:** wrap the agent-chat subscription iterator with
`wrapAsyncIteratorWithLifecycle` — emit a `keepalive` on connect and
every `APPLICATION_KEEPALIVE_INTERVAL_MS` (30s) so the connection keeps
flushing bytes and a dead connection becomes detectable.
- **Client:** track the last received event timestamp (refreshed on
every chunk/keepalive in the SSE `next` sink); new
`AgentChatStreamKeepAliveEffect` forces a resubscribe + messages refetch
after 90s of silence, so the durable Redis chunk list backfills the gap
(`firstLiveSeq` is reset on resubscribe).
### Commit 2 — stream-job idempotency + lockDuration
- Thread a `lockDuration` option through `MessageQueueWorkerOptions` +
the BullMQ driver; set `aiStreamQueue` to 10 min so long streams aren't
falsely stalled.
- Guard `StreamAgentChatJob.handle` with a `streamId`-scoped Redis lock
(`SET NX PX` + compare-and-delete release) so a stalled re-run is
skipped instead of double-processing.
## Verification
⚠️ I could **not run typecheck/lint locally** — `yarn install` could not
complete in this environment (transient registry network aborts before
the link step, so `node_modules` never populated). **Please rely on CI
for type/lint verification.** The changes are written to match existing
conventions; the points most worth a reviewer's eye are the resolver's
iterator typing and the ioredis `set(..., 'PX', ttl, 'NX')` overload.
How to confirm the root cause in prod: a frozen client with the worker
logging `StreamAgentChatJob processed in …ms` and no `[AI_CHAT_NO_TEXT]`
is the silent-death signature (check reverse-proxy idle/buffering). For
the secondary path, watch `aiStreamQueue` `stalled`/re-processed metrics
and duplicate turns around worker restarts.
## Notes / trade-offs
- The 10-min `lockDuration` means a genuinely crashed worker's job isn't
reclaimed for up to 10 min; the client-side keepalive/catch-up recovers
the view independently, and the idempotency lock prevents duplicates.
Faster dead-worker recovery could be a follow-up.
- Touches `useAgentChatSubscription.ts` / `AgentChatRuntimeEffects.tsx`
/ `stream-agent-chat.job.ts`, which #22193 also touches — trivial rebase
expected.
Opened as **draft** pending CI.
https://claude.ai/code/session_018dF82A1VcsuWMxPLmdY3dm
---
_Generated by [Claude
Code](https://claude.ai/code/session_018dF82A1VcsuWMxPLmdY3dm)_
<!-- This is an auto-generated description by cubic. -->
<a
href="https://cubic.dev/pr/twentyhq/twenty/pull/22201?utm_source=github"
target="_blank" rel="noopener noreferrer"
data-no-image-dialog="true"><picture><source
media="(prefers-color-scheme: dark)"
srcset="https://www.cubic.dev/buttons/review-in-cubic-dark.svg"><source
media="(prefers-color-scheme: light)"
srcset="https://www.cubic.dev/buttons/review-in-cubic-light.svg"><img
alt="Review in cubic"
src="https://www.cubic.dev/buttons/review-in-cubic-dark.svg"></picture></a>
<!-- End of auto-generated description by cubic. -->
## Problem
On workspaces with a sizeable AI chat history, the whole app was
freezing during navigation, including Settings (one navigation click
measured ~6.5s).
## Root cause
`AgentChatProvider` is mounted app-wide in `AppRouterProviders`, so its
effects run on every page.
On every render it would:
1. auto-select the most recently active thread
(`AgentChatThreadInitializationEffect`),
2. fetch that thread's **full message history**
(`AgentChatMessagesFetchEffect`),
3. run `AgentChatStreamingPartsDiffSyncEffect` →
`updateStreamingPartsWithDiff`, which loops over every message doing
`isDeeplyEqual(existing, incoming)` + `structuredClone`.
A large thread would produce multi-second freeze on every interaction,
app-wide. (Confirmed via a Chrome CPU profile)
## Fix
Don't run the agent-chat **message runtime** until the chat is actually
opened.
## Note
There is still room for improvement, opening AI chats would still be
very slow.
Move the browsing context out of the system prompt and injecting it
directly into the last user message instead.
Previously (before this PR), browsing context change, update system
prompt then break whole conversation history ... and caching. Now,
browsing context is sent with last message only if changed.
"Benchmark" this PR vs main :
- same conv with 3-4 turns - 60% -> 85% cache ratio || 0.31 credits ->
0.13
## PR Description
### Summary
- Add AI chat thread actions: rename, archive (soft-delete via
`deletedAt`), and hard-delete with confirmation.
- Add chat thread filtering by status (active/archived/all), group-by
mode, and last activity.
- Rework drawer/side-panel thread lists to share thread sections, item
menus, archive icons, and empty-state behavior.
- Extend server chat thread model/API with `deletedAt`, mutations,
broadcasts, and archive-aware stream guards.
### Decisions
- Two-stage lifecycle: Archive sets `deletedAt` (soft); Delete is a
separate action on archived threads that hard-deletes the row. Aligns
with Twenty's soft-delete convention (Felix's suggestion).
- `lastMessageAt` is derived from `MAX(agentMessage.createdAt)` on read,
not stored. List query does inline aggregation for sort; `@ResolveField`
covers single-thread / mutation paths so the schema contract is honest
everywhere. Matches `timeline-messaging.service.ts` precedent and the
existing `totalInputCredits` / `totalOutputCredits` `@ResolveField`
pattern in the same resolver.
- Replaced auto-CRUD `chatThreads` (cursor-paginated Connection) with a
custom `[AgentChatThreadDTO!]` resolver. Frontend metadata-store treats
threads as a flat collection and filters/sorts client-side, so cursor
pagination was performative.
- Sending in an archived chat unarchives it optimistically on the client
and authoritatively on the server.
- Grouping and last-activity filtering use `lastMessageAt ?? updatedAt`
so archive/rename don't bump threads in the list.
- Kept metadata-store core API unchanged; AI chat uses the same local
cast pattern already used by other metadata-store partial updates.
https://github.com/user-attachments/assets/1b179b7b-1a2a-4a7a-aa0a-c88f6f051a87
## Summary
The "AI" acronym was rendered inconsistently across the codebase. The
backend AI module had settled on PascalCase `Ai` (`AiAgentModule`,
`AiBillingService`, `AiChatModule`, `AiModelRegistryService`, etc.),
while frontend components, several DTOs, a few types, and shared
identifiers still used all-caps `AI` (`AIChatTab`,
`AISystemPromptPreviewDTO`, `SettingsPath.AIPrompts`, ...). CLAUDE.md
specifies PascalCase for classes; this PR normalizes everything internal
to `Ai`.
**This is a pure internal rename.** The GraphQL schema is untouched —
`@ObjectType` decorator string arguments, resolver method names (which
become Query/Mutation field names), gql template contents, and the
`generated-metadata/graphql.ts` file are preserved verbatim. The only
visible change is TypeScript identifiers and file names.
## Also folded in (adjacent cleanups)
- **`AgentModelConfigService` → `AiModelConfigService`**. Lives in
`ai-models/` and is used by multiple AI code paths, not just the Agent
entity. The "Agent" prefix was misleading.
- **`generate-text-input.dto.ts` → `generate-text.input.ts`**. The
`ai-agent/dtos/` folder already uses `<entity>.input.ts` convention for
Input classes (`create-agent.input.ts` etc.); the old path mixed
`.dto.ts` file extension with a class that has no DTO suffix. File
rename only; class stays `GenerateTextInput`.
- **Removed stale TODO** in `ai-model-config.type.ts` that asked for the
`AiModelConfig` rename that this PR performs.
## Rename methodology
Bulk rename via perl with anchored regex
`(?<!['"])(?<![A-Z.])AI([A-Z])(?=[a-z])/Ai$1/g`:
- **Lookbehind for non-uppercase** skips adjacent acronyms (`MOSAIC`,
`OIDCSSO`) and leaves `AIRBNB_ID` alone.
- **Lookbehind for non-quote** protects most string literals.
- **Lookahead for lowercase** restricts matches to PascalCase
identifiers (`AIChatTab`), leaving SCREAMING_SNAKE constants untouched.
Strict file-scope exclusions: `generated-metadata/**`, `generated/**`,
`locales/**`, `migrations/**`, `illustrations/**`, `halftone/**`, and
the two gql template files (`queries/getAISystemPromptPreview.ts`,
`mutations/uploadAIChatFile.ts`).
Post-rename reverts for identifiers where the regex was too eager:
- Backend resolver method names kept: `getAISystemPromptPreview`,
`uploadAIChatFile` (they are GraphQL field names).
- `@ObjectType('AdminAIModels')` / `('AISystemPromptPreview')` /
`('AISystemPromptSection')` kept as-is.
- Backend classes `ClientAIModelConfig` / `AdminAIModelConfig` kept
as-is (they use `@ObjectType()` with no argument, so the class name IS
the schema name).
- External-library symbols restored: `OpenAIProvider`,
`createOpenAICompatible`, `vercelAIIntegration`.
File renames use a two-step rename to work on macOS case-insensitive
filesystems: `git mv X.tsx X.tsx.tmp && git mv X.tsx.tmp renamed.tsx`.
## Diff audit
- 0 changes to migrations
- 0 changes to locale `.po` / `.ts` files
- 0 changes to `generated-metadata/graphql.ts`
- 0 changes to website illustration files (base64 blobs preserved)
- 0 renames inside user-facing translation strings (`t\`…\``,
`msg\`…\``, `<Trans>…</Trans>`)
## Test plan
- [x] `npx nx typecheck twenty-server` — PASS
- [x] `npx nx typecheck twenty-front` — PASS
- [x] `npx jest ai-model admin agent-role` — 79/79 PASS
- [x] `npx oxlint --type-aware` on 118 changed files — 0 errors
- [x] `npx prettier --check` on 118 changed files — clean
- [ ] CI
## What changed
This refactor fixes AI chat error surfacing by aligning both the backend
and frontend with the existing GraphQL error architecture instead of
adding AI-local error translation.
On the backend:
- add a dedicated GraphQL billing exception path
- register billing GraphQL handling globally for GraphQL requests
- reuse the existing AI GraphQL interceptor path for agent/chat
exceptions
- keep billing status classification shared between REST and GraphQL
- remove the earlier attempt to preserve `CustomException` metadata in
the global GraphQL fallback
On the frontend:
- keep the original Apollo GraphQL error object in AI chat state
- reuse shared Apollo/GraphQL helpers for user-facing messages and
error-type checks
- delete AI-specific error extraction helpers that duplicated generic
GraphQL parsing
- replace a few direct `extensions.subCode` call sites with a shared
predicate
## Why it changed
The original bug was that `BillingException` and AI exceptions thrown
from chat were not being translated into GraphQL errors with the
expected `extensions.subCode` and `extensions.userFriendlyMessage`, so
the AI chat UI had nothing structured to inspect.
An intermediate fix worked mechanically but pushed `CustomException`
handling into the global GraphQL fallback, which blurred the intended
layering. This PR moves the behavior back to explicit GraphQL edges.
## Root cause
`AgentChatResolver` could throw `BillingException` and `AgentException`,
but:
- billing had a REST exception filter and no shared GraphQL equivalent
- AI chat was not consistently using the same GraphQL exception
translation path as the sibling AI resolver
- the frontend chat UI had drifted into AI-specific error parsing
instead of consuming the same structured Apollo errors as the rest of
the app
## Impact
- `BILLING_CREDITS_EXHAUSTED` is now preserved through GraphQL and can
render the existing credits-exhausted UI in chat
- `API_KEY_NOT_CONFIGURED` is preserved through the AI GraphQL path
- AI chat now follows the same general GraphQL error consumption pattern
as the rest of the frontend
- billing GraphQL handling is less dependent on individual resolver
authors remembering to add a filter
## Validation
- `yarn jest --config packages/twenty-server/jest.config.mjs
packages/twenty-server/src/engine/core-modules/billing/utils/__tests__/billing-graphql-api-exception-handler.util.spec.ts
packages/twenty-server/src/engine/metadata-modules/ai/ai-agent/utils/__tests__/agent-graphql-api-exception-handler.util.spec.ts`
- `yarn jest --config packages/twenty-front/jest.config.mjs
packages/twenty-front/src/utils/__tests__/is-graphql-error-of-type.util.test.ts`
- `npx oxlint --type-aware ...` on touched backend/frontend files
- `npx prettier --check ...` on touched backend/frontend files
## Follow-up ideas
- consolidate frontend GraphQL error helpers further so more existing
direct `extensions.subCode` checks move to shared utilities
- consider whether common GraphQL exception filter registration should
live in a more explicit GraphQL-specific module instead of
`CoreEngineModule`
- add an end-to-end test for a real `sendChatMessage` GraphQL failure
path in AI chat
---------
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
## Summary
Refactored AI chat state management to use component family states
instead of global atoms. This change enables proper state isolation per
chat thread, preventing state leakage between different chat instances.
## Key Changes
- **Converted global atoms to component family states:**
- `agentChatErrorState` → `agentChatErrorComponentFamilyState`
- `agentChatIsStreamingState` →
`agentChatIsStreamingComponentFamilyState`
- `agentChatFirstLiveSeqState` →
`agentChatFirstLiveSeqComponentFamilyState`
- `agentChatHandleEventCallbackState` →
`agentChatHandleEventCallbackComponentFamilyState`
- `agentChatUsageState` → `agentChatUsageComponentFamilyState`
- `currentAIChatThreadTitleState` →
`currentAIChatThreadTitleComponentFamilyState`
- **Updated state access patterns:**
- Introduced `useAtomComponentFamilyStateCallbackState` hook to get
family state callbacks
- Changed from direct atom access to family key-based access using `{
threadId }` as the family key
- Updated all components and hooks to use the new family state patterns
- **Removed instance ID dependency:**
- Eliminated `AGENT_CHAT_INSTANCE_ID` constant usage
- Simplified state access by using thread ID directly as the family key
- **Updated affected components and hooks:**
- `useAgentChatSubscription`: Now creates family state atoms per thread
- `AgentChatMessagesFetchEffect`: Uses family state callbacks for event
handling
- `AgentChatThreadInitializationEffect`: Sets family state values per
thread
- `useAIChatThreadClick`: Updates family states when switching threads
- `AIChatErrorUnderMessageList`, `AIChatLastMessageWithStreamingState`,
`AIChatEmptyState`, `AIChatStandaloneError`, `SendMessageButton`,
`AIChatContextUsageButton`, `SidePanelAskAIInfo`: Updated to use family
state values with thread ID
## Implementation Details
- All new component family states use
`AgentChatComponentInstanceContext` for proper component-level isolation
- Family key structure: `{ threadId: string | null }`
- Removed `disposed` flag logic as it's no longer needed with proper
state isolation
- Updated dependency arrays in hooks to include family callback
functions
https://claude.ai/code/session_01D8syiJtCXu5bEHSr5KYkCt
---------
Co-authored-by: Claude <noreply@anthropic.com>
## Summary
This PR refactors the AI chat thread state management to use `null`
instead of a sentinel string value (`AGENT_CHAT_UNKNOWN_THREAD_ID`) to
represent an uninitialized or new chat thread. This improves type safety
and makes the code more idiomatic by using `null` to represent the
absence of a value.
## Key Changes
- **Removed sentinel constant**: Deleted `AGENT_CHAT_UNKNOWN_THREAD_ID`
constant and replaced all usages with `null`
- **Updated state types**: Changed `currentAIChatThreadState`,
`agentChatLastDiffSyncedThreadState`, and
`agentChatDisplayedThreadState` to use `string | null` type with `null`
as default value
- **Updated component family states**: Modified message-related state
families to accept `threadId: string | null` instead of `threadId:
string`
- **Refined null checks**: Added explicit `null` checks in:
- `AgentChatMessagesFetchEffect`: Updated `isNewThread` logic to check
for `null` first
- `useAIChatThreadClick` and `useSwitchToNewAIChat`: Added guards to
only save drafts when `currentAIChatThread !== null`
- `AgentChatThreadInitializationEffect`: Added null check before UUID
validation
- `useEnsureAgentChatThreadIdForSend`: Added null check before comparing
with draft key
- **Updated fallback logic**: Used nullish coalescing operator (`??`) in
`AIChatTab` and `useAIChatEditor` to default to
`AGENT_CHAT_NEW_THREAD_DRAFT_KEY` when thread is null
- **Enhanced refetch safety**: Added early return in
`handleRefetchMessages` to prevent refetching when in new thread state
## Implementation Details
- The change maintains backward compatibility by treating `null` the
same way the code previously treated `AGENT_CHAT_UNKNOWN_THREAD_ID`
- All draft saving operations now safely check for null before
attempting to store drafts
- The nullish coalescing pattern (`currentAIChatThread ??
AGENT_CHAT_NEW_THREAD_DRAFT_KEY`) ensures proper fallback behavior when
accessing draft storage
https://claude.ai/code/session_01Pz8KCygSNgBPYsndbMq8f7
---------
Co-authored-by: Claude <noreply@anthropic.com>
## Summary
- **Queue messages while streaming**: Messages sent during active AI
streaming are queued server-side and auto-flushed when the current
stream completes. Frontend renders queued messages optimistically in a
dedicated queue UI.
- **Drop `@ai-sdk/react` + `resumable-stream`**: Replace the dual HTTP
SSE + AI SDK client architecture with a single GraphQL SSE subscription
per thread. All events (token chunks, message persistence, queue
updates, errors) flow through Redis PubSub → GraphQL subscription.
- **Server-driven architecture**: The server decides whether to queue or
stream (via `POST /:threadId/message`). The frontend mirrors this
decision for optimistic rendering but defers to the server response.
- **Reuse AI SDK accumulation logic**: `readUIMessageStream` from the
`ai` package handles chunk-to-message accumulation on the frontend,
avoiding a custom 780-line accumulator.
## Key files
**Backend:**
- `agent-chat-event-publisher.service.ts` — publishes events to Redis
PubSub
- `agent-chat-subscription.resolver.ts` — GraphQL subscription resolver
- `stream-agent-chat.job.ts` — publishes chunks via PubSub instead of
resumable-stream
- `agent-chat.controller.ts` — unified `POST /:threadId/message`
endpoint
**Frontend:**
- `useAgentChatSubscription.ts` — subscribes to `onAgentChatEvent`,
bridges to `readUIMessageStream`
- `useAgentChat.ts` — send/stop/optimistic rendering (no more AI SDK)
- `AgentChatStreamSubscriptionEffect.tsx` — replaces
`AgentChatAiSdkStreamEffect.tsx`
## Test plan
- [ ] Send message on new thread → optimistic render, streaming response
appears
- [ ] Send message while streaming → queued instantly (no flash in main
thread)
- [ ] Queued message auto-flushes after current stream completes
- [ ] Remove queued message via queue UI
- [ ] Stop streaming mid-response
- [ ] Leave chat idle for several minutes → streaming still works after
(SSE client recycling)
- [ ] Token refresh during session → requests succeed (authenticated
fetch)
- [ ] Switch threads while streaming → clean subscription handoff
Made with [Cursor](https://cursor.com)
---------
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
This PR:
- Breaks useAgentChatData into focused effect components (streaming,
fetch,
init, auto-scroll, diff sync)
- Splits message list into non-last (stable) + last (streaming/error) to
prevent full re-renders on each stream chunk
- Adds scroll-to-bottom button and MutationObserver-based auto-scroll on
thread switch
- Lifts loading state from context to atoms
- Adds areEqual to selector
factories
We could improve further but this sets up a robust architecture for
further refactoring.
## Messages flow
The flow of messages loading and streaming is now more solid.
Everything goes out from `AgentChatAiSdkStreamEffect`, whether loaded
from the DB or streaming directly, and every consumers is using only one
atom `agentChatMessagesComponentFamilyState`
## Data sync effect with callbacks new hook
See
`packages/twenty-front/src/modules/apollo/hooks/useQueryWithCallbacks.ts`
which allows to fix Apollo v4 migration leftovers and is an
implementation of the pattern we talked about with @charlesBochet
We could refine this pattern in another PR.
# Before
https://github.com/user-attachments/assets/84e7a96f-6790-405d-8a73-2dacbf783be5
# After
https://github.com/user-attachments/assets/4c692e3a-2413-4513-abcc-44d0da311203
Co-authored-by: Charles Bochet <charles@twenty.com>
This PR adds the necessary tool to create a demo workspace with :
relevant custom objects and fields, mock data and a real dashboard with
graph widgets.
It is still a bit under-optimized and slow but it works.
This PR also adds an AI tool that allows to see what happens in real
time, it navigates the app and waits when necessary.
---------
Co-authored-by: Félix Malfait <felix.malfait@gmail.com>
- Remove feature flag
- Remove legacy methods in file-upload and file-service
- Migrate AI Chat to new file management
---------
Co-authored-by: Charles Bochet <charles@twenty.com>
## Summary
Removes the `recoil` dependency entirely from `package.json` and
`twenty-front/package.json`, completing the migration to Jotai as the
sole state management library.
Removes all Recoil infrastructure: `RecoilRoot` wrapper from `App.tsx`
and test decorators, `RecoilDebugObserver`, Recoil-specific ESLint rules
(`use-getLoadable-and-getValue-to-get-atoms`,
`useRecoilCallback-has-dependency-array`), and legacy Recoil utility
hooks/types (`useRecoilComponentState`, `useRecoilComponentValue`,
`createComponentState`, `createFamilyState`, `getSnapshotValue`,
`cookieStorageEffect`, `localStorageEffect`, etc.).
Renames all `V2`-suffixed Jotai state files and types to their canonical
names (e.g., `ComponentStateV2` -> `ComponentState`,
`agentChatInputStateV2` -> `agentChatInputState`, `SelectorCallbacksV2`
-> `SelectorCallbacks`), and removes the now-redundant V1 counterparts.
Updates ~433 files across the codebase to use the renamed Jotai imports,
remove Recoil imports, and clean up test wrappers (`RecoilRootDecorator`
-> `JotaiRootDecorator`).
## Summary
Replaces the static "Ask AI" header in the command menu with the
conversation’s auto-generated title once it’s set after the first
message.
## Changes
- **Backend:** Title is generated after the first user message (existing
behavior).
- **Frontend:** After the first stream completes, we fetch the thread
title and sync it to:
- `currentAIChatThreadTitleState` (persists across command menu
close/reopen)
- Command menu page info and navigation stack (so the title survives
back navigation)
- **Entry points:** Opening Ask AI from the left nav or command center
uses the same title resolution (explicit `pageTitle` → current thread
title → "Ask AI" fallback).
- **Race fix:** Title sync only runs when the thread that finished
streaming is still the active thread, so switching threads mid-stream
doesn’t overwrite the current thread’s title.
---------
Co-authored-by: Félix Malfait <felix@twenty.com>
Co-authored-by: Cursor <cursoragent@cursor.com>
⚠️ **AI-generated PR — not ready for review** ⚠️
cc @FelixMalfait
---
## Changes
### System prompt improvements
- Explicit skill-before-tools workflow to prevent the model from calling
tools without loading the matching skill first
- Data efficiency guidance (default small limits, use filters)
- Pluralized `load_skill` → `load_skills` for consistency with
`load_tools`
### Token usage reduction
- Output serialization layer: strips null/undefined/empty values from
tool results
- Lowered default `find_*` limit from 100 → 10, max from 1000 → 100
### System object tool generation
- System objects (calendar events, messages, etc.) now generate AI tools
- Only workflow-related and favorite-related objects are excluded
### Context window display fix
- **Bug**: UI compared cumulative tokens (sum of all turns) against
single-request context window → showed 100% after a few turns
- **Fix**: Track `conversationSize` (last step's `inputTokens`) which
represents the actual conversation history size sent to the model
- New `conversationSize` column on thread entity with migration
### Workspace AI instructions
- Support for custom workspace-level AI instructions
---------
Co-authored-by: claude[bot] <41898282+claude[bot]@users.noreply.github.com>
## Summary
Fixes an issue where the AI chat would show loading shimmers
indefinitely when opened on a workspace with no conversation history.
**Root cause:** The `isLoading` state in `useAgentChat` included
`!currentAIChatThread`. On workspaces with no chat threads,
`currentAIChatThread` remained `null`, causing `isLoading` to be
permanently `true`.
**Changes:**
- Remove `!currentAIChatThread` from `isLoading` calculation in
`useAgentChat` - this state should only reflect streaming/file selection
status
- Auto-create a chat thread in `useAgentChatData` when the threads query
returns empty, ensuring a valid thread exists for the `useChat` hook to
initialize properly
- Add primary font color to empty state title for better visibility
## Test plan
1. Create a new workspace or use a workspace with no AI chat history
2. Open the AI chat
3. Verify the empty state shows (not infinite loading shimmer)
4. Send a message and verify it works correctly
Made with [Cursor](https://cursor.com)
---------
Co-authored-by: Cursor <cursoragent@cursor.com>
## 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
- 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
## Summary
This PR introduces a comprehensive agent evaluation system and refactors
the AI module structure for better organization.
## Key Changes
### 🎯 Agent Evaluation System
- Added **Agent Turn Evaluation** entities, DTOs, and database schema
- New GraphQL mutations: `evaluateAgentTurn` and `runEvaluationInput`
- Added `evaluationInputs` field to Agent entity for storing test inputs
- New `AgentTurnGraderService` for automatic turn evaluation
- Added evaluation UI with new **Evals** and **Logs** tabs in agent
detail pages
### 🏗️ Entity & Module Refactoring
- Renamed `AgentChatMessage` → `AgentMessage` for clarity
- Consolidated chat entities: `AgentMessage`, `AgentTurn`, and
`AgentChatThread`
- Reorganized AI modules under `ai/` subdirectory structure
- Updated imports across codebase to reflect new module paths
### 🤖 New Agents & Roles
- Added **Dashboard Builder Agent** for dashboard creation and
management
- Added **Dashboard Manager Role** with appropriate permissions
- Updated role permissions to be more granular (users vs agents vs API
keys)
### 🔐 Permission System Updates
- Added `HTTP_REQUEST_TOOL` permission flag
- Updated Workflow Manager role permissions (restricted tool access)
- Enhanced permission flag types to differentiate between user/agent/API
key contexts
- Added `isRelevantForAgents`, `isRelevantForApiKeys`,
`isRelevantForUsers` to permission flags
### 📨 Message Role Enhancement
- Added `system` role to `AgentMessageRole` enum (alongside
user/assistant)
- Updated message handling to support system prompts
### 🎨 UI/UX Improvements
- New tabs in agent detail: **Evals** and **Logs**
- Added turn detail page: `/ai/agents/:agentId/turns/:turnId`
- Fixed text overflow in `SettingsListItemCardContent`
- Updated role applicability labels ("Assignable to Workspace Members")
### 🛠️ Technical Improvements
- Fixed Zod schema validation for UUID and Date fields (use string
validators)
- Updated `ToolRegistryService` to properly register HTTP tool with
permission flag
- Enhanced error handling in agent execution services
- Updated database migrations for new entity schema
## Database Migrations
- `1764210000000-add-system-role-to-agent-message.ts`
- `1764220000000-add-evaluation-inputs-to-agent.ts`
- `1764200000000-add-agent-turn-evaluation.ts`
- `1764100000000-refactor-agent-chat-entities.ts`
## Testing
- [ ] Agent evaluation flow tested
- [ ] Dashboard Builder agent tested
- [ ] Permission system validated
- [ ] UI tabs and navigation tested
- [ ] Database migrations run successfully
## Breaking Changes
⚠️ **Entity Rename**: `AgentChatMessage` renamed to `AgentMessage` -
GraphQL queries need updating
## Related Issues
<!-- Link any related issues here -->
## Screenshots
<!-- Add screenshots if applicable -->
Closes [#1583](https://github.com/twentyhq/core-team-issues/issues/1583)
- Removed `messageId` column from `File` table and its references in
code (unrelated to this PR)
- Updated AI chat to use React Context API with persistent provider in
`CommandMenuContainer`
- Converted all AI chat component states to regular Recoil atoms (no
instance context needed)
---------
Co-authored-by: Félix Malfait <felix@twenty.com>
This PR removes any V2 naming in state management logic and some minor
utils and hooks.
It has a lot of changes but nearly all of them were made by the rename
functionality of vscode which is deterministic, so it shouldn't
introduce any regression.
QA has been made on this PR on the main features of the app without any
noticeable issue.
Also renamed some other v2 naming related items :
- TextInputV2 => TextInput
- TextInput => SettingsTextInput
- ObjectFilterDropdownFilterSelectMenuItemV2 =>
ObjectFilterDropdownFilterSelectMenuItem
- useInitDraftValueV2 => useInitDraftValue
- useOpenRecordTableCellV2 => useOpenRecordTableCell
## TODO
- [ ] add dropdown to use records from outside the context
- [x] add loader for files chip
- [x] add roleId where it's necessary
- [x] Split AvatarChip in two components. One with the icon that will
call the second with leftComponent.
- [ ] Fix tests
- [x] Fix UI regression on Search
Félix Malfait
