🎚️ feat: Per-User Skill Active/Inactive Toggle with Ownership-Aware Defaults - #12692
Merged
Merged
Conversation
Collaborator
Author
|
@codex review |
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 624370238b
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
| const isOwned = s.author.toString() === userId; | ||
| return isOwned ? true : defaultActiveOnShare; | ||
| }) | ||
| : skills; |
There was a problem hiding this comment.
This code fails open when userId is absent by returning all accessible skills, which bypasses per-user active/inactive overrides. I checked the repo and there are active agent entry points (for example api/server/controllers/agents/openai.js and api/server/controllers/agents/responses.js) that still call initializeAgent without passing userId/skillStates, so disabled skills can still be injected in those flows. Make this path fail closed (or require userId whenever accessibleSkillIds are provided) so the toggle is enforced consistently.
Useful? React with 👍 / 👎.
Comment on lines
+56
to
+59
| if (typeof key !== 'string' || key.length === 0 || key.length > MAX_KEY_LENGTH) { | ||
| return res.status(400).json({ | ||
| message: `Each skill ID must be a non-empty string (max ${MAX_KEY_LENGTH} chars)`, | ||
| }); |
There was a problem hiding this comment.
The validation only checks key length/type, but Mongoose Map keys reject characters like . and leading $; those payloads will throw during updateUser and hit the generic 500 handler instead of returning a client 400. This turns malformed user input into server errors and makes the endpoint brittle. Add explicit key-pattern validation (or map-cast error translation) before persisting.
Useful? React with 👍 / 👎.
Comment on lines
+83
to
+84
| const next = { ...skillStates, [skill._id]: !current }; | ||
| save(next); |
There was a problem hiding this comment.
The toggle writes a full replacement object (next) derived from the current in-memory skillStates; before the initial query resolves that state is the empty fallback, so an early click can overwrite existing server-side overrides with only one key. Because the API replaces the entire map, this can silently drop prior toggles. Disable toggling while the initial fetch is loading or merge against the latest cached/server value before posting.
Useful? React with 👍 / 👎.
Collaborator
Author
|
@codex review |
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 4d0e6f3614
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
| if (!userId) { | ||
| return false; | ||
| } | ||
| return s.author.toString() === userId ? true : defaultActiveOnShare; |
There was a problem hiding this comment.
This fallback makes shared skills inactive whenever injectSkillCatalog is called without the new skillStates/defaultActiveOnShare context, and some existing agent entrypoints still only provide accessibleSkillIds (for example the OpenAI/Responses controllers). In those paths, user toggles and admin defaultActiveOnShare are silently ignored, so shared skills cannot be activated even though they are accessible; require these params at call sites or bypass ownership filtering when they are absent.
Useful? React with 👍 / 👎.
Comment on lines
+349
to
+353
| <SkillToggle | ||
| enabled={isSkillEnabled} | ||
| onChange={handleToggleEnabled} | ||
| ariaLabel={toggleAriaLabel} | ||
| /> |
There was a problem hiding this comment.
The new SkillToggle is focusable inside a row that already handles Enter/Space to navigate, so keyboard events from the toggle bubble up and trigger handleSkillClick. That means keyboard users can be navigated away when trying to flip the switch (and the toggle action may be canceled), making the active/inactive control unreliable from the list; isolate key handling to the row itself or stop propagation for toggle key activation.
Useful? React with 👍 / 👎.
Merged
Collaborator
Author
|
@codex review |
There was a problem hiding this comment.
💡 Codex Review
https://github.com/danny-avila/LibreChat/blob/c5442a74f850df62b3362752f235cc21dcc7558e/packages/api/src/agents/skills.ts#L68-L71
injectSkillCatalog applies SKILL_CATALOG_LIMIT when fetching accessible skills, then filters that page by active/inactive state. When a user has more than 100 accessible skills and many recent ones are shared (default inactive), active owned skills past the first page are never considered, so the agent can get an empty or truncated catalog even though active skills exist. Apply the limit after active-state filtering (or page until the active quota is filled) so inactive rows do not consume the catalog budget.
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
Collaborator
Author
|
@codex review |
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 9a3f988b8b
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
Comment on lines
+96
to
+98
| if (getQuery.isLoading) { | ||
| return; | ||
| } |
There was a problem hiding this comment.
The toggle guard only checks getQuery.isLoading, so after an initial GET /skills/active failure on a cold load, skillStates falls back to {} and the next toggle posts a partial map built from an empty baseline. Because the backend updateSkillStatesController replaces the full skillStates map, this can wipe previously saved overrides for other skills when the POST succeeds after a transient GET error.
Useful? React with 👍 / 👎.
Comment on lines
+61
to
+64
| const queueRef = useRef<{ | ||
| pending: TSkillStatesResponse | null; | ||
| inFlight: boolean; | ||
| }>({ pending: null, inFlight: false }); |
There was a problem hiding this comment.
The request queue is stored in a component-local useRef, but this hook is used in both SkillList and SkillDetail, so each instance can send concurrent full-map updates independently. If two toggles are made before the first network round-trip completes (for example one in the list and one in detail), whichever POST finishes last overwrites the other change, dropping user toggles due to last-writer-wins on the whole skillStates object.
Useful? React with 👍 / 👎.
Collaborator
Author
|
@codex review |
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 3966bcb6d5
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
| if (!canToggle) { | ||
| return; | ||
| } | ||
| const baseline = writeQueue.pending ?? skillStates; |
There was a problem hiding this comment.
When a toggle is already in flight, this line falls back to the hook’s captured skillStates snapshot, which can be stale until React Query re-renders after onMutate. If the user toggles another skill before that re-render (e.g., quickly from list/detail views), the second payload is built from an old map and can drop the first toggle; because the backend POST /skills/active replaces the full map, this causes lost updates rather than just temporary UI drift.
Useful? React with 👍 / 👎.
…ults - Add `skillStates` map (Record<string, boolean>) to user schema for per-user active/inactive overrides on skills - Add `defaultActiveOnShare` to interface.skills config (default: false) so admins can control whether shared skills auto-activate - Add GET/POST /api/user/settings/skills/active endpoints with validation - Add React Query hooks with optimistic mutations for skill states - Add useSkillActiveState hook with ownership-aware resolution: owned skills default active, shared skills default inactive - Add toggle switch UI to SkillListItem and SkillDetail components - Filter inactive skills in injectSkillCatalog before agent injection - Add localization keys for active/inactive labels
Mongoose .lean() flattens Map to a plain object, causing type incompatibility with IUser in methods that return lean documents.
- Fail-closed when userId is absent: filter rejects all shared skills instead of passing them through unfiltered (Codex P1) - Validate Mongoose Map key characters (reject . and $) in controller to return 400 instead of a 500 from schema validation (Codex P2) - Block toggle while initial skill states query is loading to prevent overwriting server-side overrides with an empty snapshot (Codex P2) - Extract shared SkillToggle component, eliminating duplicate toggle markup in SkillListItem and SkillDetail (Finding #3) - Move skill state query/mutation hooks from Favorites.ts to Skills/queries.ts per feature-directory convention (Finding #4) - Fix hardcoded English aria-label in SkillListItem by passing the localized string from the parent SkillList (Finding #5) - Fix inline arrow in SkillList render loop: pass stable callback reference so SkillListItem memo() is not invalidated (Finding #1) - Extract toRecord() helper in controller to DRY the Map-to-Object conversion (Finding #6) - Remove Promise.resolve wrapping synchronous config read (Finding #8) - Remove unused TUpdateSkillStatesRequest type (Finding #12)
The original inline toggle had tabIndex={-1} so the row itself
remained the sole tab target. The extraction into SkillToggle
dropped this prop, making every list toggle a tab stop. Add an
optional tabIndex prop and pass -1 from SkillListItem.
- Add skillStates/defaultActiveOnShare loading to openai.js and responses.js controllers so shared-skill activation is respected across all agent entry points, not just initialize.js (Codex P1) - Stop keydown propagation on SkillToggle so Enter/Space does not bubble to the parent row's navigation handler (Codex P2)
- Paginate listSkillsByAccess (up to 10 pages of 100) until the active catalog quota is filled, so inactive shared skills in recent positions do not starve active owned skills past the first page (Codex P1) - Extend listSkillsByAccess interface with cursor/has_more/after for catalog pagination - Serialize skill-state writes via a ref queue: one in-flight request at a time, with the latest desired state sent when the previous one settles. Prevents last-response-wins races where an older request overwrites newer toggles (Codex P2)
…rror - Move the write queue from a per-instance useRef to a module-scoped object so every mount of useSkillActiveState (SkillList, SkillDetail, etc.) serializes against the same in-flight slot. Prior per-instance queues allowed two components to race full-map POSTs (Codex P1) - Extend the toggle guard beyond isLoading: also block when isError is true or data is undefined. Prevents a failed GET from seeding a toggle with an empty baseline that would wipe server-side overrides on the next successful POST (Codex P1)
- Read toggle baseline from React Query cache via queryClient.getQueryData instead of the captured skillStates closure. The closure can be stale between onMutate's setQueryData and the next render, so rapid successive toggles would build on old state and drop earlier changes (Codex P1) - Surface the MAX_SKILL_STATES_EXCEEDED error code with a specific toast key (com_ui_skill_states_limit) so users understand the 200-cap rather than seeing a generic error - Prune orphaned entries (skillIds whose Skill doc no longer exists) on both GET and POST in SkillStatesController. Self-heals over time without needing cascade-delete hooks or a migration job. Uses one indexed Skill._id query per request
danny-avila
force-pushed
the
feat/skill-active-states
branch
from
April 16, 2026 22:22
5034f71 to
1366671
Compare
Collaborator
Author
|
@codex review |
Extract the active-state resolution logic from a closure inside injectSkillCatalog into an exported resolveSkillActive helper, then cover every branch of the precedence matrix: - Fails closed when userId is absent (even with defaultActiveOnShare=true) - Explicit override wins over ownership and config (both true and false) - Owned skills default to active when no override is set - Shared skills default to defaultActiveOnShare value - Undefined skillStates behaves identically to an empty object - defaultActiveOnShare defaults to false when omitted - Owned skills ignore defaultActiveOnShare entirely Closes Finding #2 from the pre-rebase comprehensive review. Mirrors the existing scopeSkillIds test style; injectSkillCatalog now calls resolveSkillActive instead of inlining the closure.
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 1366671ecc
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
| override !== undefined | ||
| ? override | ||
| : resolveDefault(skill.author, userId, defaultActiveOnShare); | ||
| writeQueue.pending = { ...baseline, [skill._id]: !currentActive }; |
There was a problem hiding this comment.
The toggle path always writes skillId -> !currentActive into the map, even when that new value equals the ownership/config default. Because the backend enforces a hard cap of 200 entries (MAX_SKILL_STATES), users can exhaust the quota by toggling many skills on and then back to their default state, after which further toggles fail with MAX_SKILL_STATES_EXCEEDED despite having few (or no) meaningful overrides. The toggle logic should delete entries when the new value matches the resolved default instead of persisting a redundant override.
Useful? React with 👍 / 👎.
- Remove the per-row toggle from SkillListItem and the active-state plumbing (hook call, isSkillEnabled/onToggleEnabled/toggleAriaLabel props) from SkillList. The detail view is now the single place to change a skill's active state - Drop dim/muted styling for inactive skills in the sidebar: without a control there, the visual indication has nowhere to land - Resize SkillToggle to match neighbor buttons: outer h-9 container, h-6 w-11 track with size-5 knob, no label span. The 'Active' / 'Inactive' text that accompanied the detail-view toggle is removed - Remove the now-unused label prop and tabIndex prop (the tabIndex existed only for the list-row context) from SkillToggle. Drop the onKeyDown stopPropagation for the same reason - Remove now-orphaned com_ui_skill_active / com_ui_skill_inactive translation keys
Container stays at h-9 to match neighbor button heights. The toggle track itself drops from h-6 w-11 to h-5 w-9, with a size-4 knob travelling 1.125rem on activation. Visually lighter inside the row.
…ault
When a toggle lands on the ownership/config default, delete the key
from the map instead of persisting `{id: defaultValue}`. Without this,
a user toggling a skill off and back on would leave `{id: true}` for
an owned skill (whose default is already true), silently consuming a
slot against the 200-entry cap. Repeated round-trip toggles could
exhaust the quota with zero meaningful overrides (Codex P2).
Preserves the exceptions-list invariant that the runtime-resolution
design depends on.
Collaborator
Author
|
@codex review |
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: d57997e4dd
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
Comment on lines
+24
to
+26
| const ids = Object.keys(skillStates).filter((id) => isValidObjectIdString(id)); | ||
| if (ids.length === 0) { | ||
| return skillStates; |
There was a problem hiding this comment.
pruneOrphans returns the input map unchanged when ids.length === 0, so a payload containing only non-ObjectId keys bypasses the cleanup path entirely and gets persisted as user state. This contradicts the endpoint’s “self-heal orphaned overrides” behavior and lets malformed keys accumulate in skillStates (and count toward the 200-entry cap) until a later write happens to include at least one valid ObjectId.
Useful? React with 👍 / 👎.
Comment on lines
+74
to
+79
| if (entries.length > MAX_SKILL_STATES) { | ||
| return res.status(400).json({ | ||
| code: 'MAX_SKILL_STATES_EXCEEDED', | ||
| message: `Maximum ${MAX_SKILL_STATES} skill state overrides allowed`, | ||
| limit: MAX_SKILL_STATES, | ||
| }); |
There was a problem hiding this comment.
The max-size guard runs on the raw payload before orphan cleanup, but orphan pruning is done later in the same request. If a user is near the 200-entry cap and some saved overrides now point to deleted skills, adding a new toggle can incorrectly return MAX_SKILL_STATES_EXCEEDED even though pruning would bring the stored map back under the limit.
Useful? React with 👍 / 👎.
Reorder the update controller so pruneOrphans runs before the 200-cap check. Without this, a user near the cap with some orphaned entries (skills deleted since their last GET) could send a payload that would pass after pruning but gets rejected by the raw-size check first. Add a sanity cap on raw payload size (2 * MAX_SKILL_STATES) so abusive inputs do not reach the DB query, and enforce the real cap on the pruned result instead. Harden pruneOrphans: the earlier early-return path could pass non-ObjectId keys through unchanged. Now only valid ObjectIds are returned, and the Skill-model-unavailable fallback filters by format. Also add isValidObjectIdString validation at the input boundary so malformed (but otherwise non-Mongo-unsafe) keys never reach persistence (Codex P2 x2).
Collaborator
Author
|
@codex review |
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 01caa85960
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
| } | ||
|
|
||
| return { toolDefinitions, skillCount: skills.length }; | ||
| return { toolDefinitions, skillCount: activeSkills.length }; |
There was a problem hiding this comment.
injectSkillCatalog now filters to activeSkills, but it only uses that set to build the prompt catalog and returns no filtered ID list. The runtime executor still authorizes skill tool calls against the full accessibleSkillIds set (initialize.ts keeps the unfiltered IDs, and handlers.ts resolves getSkillByName from that set), so a deactivated skill can still be invoked by name whenever at least one active skill keeps the tool registered. This breaks the new active/inactive guarantee and allows disabled skills to run in production conversations.
Useful? React with 👍 / 👎.
…pe queue per user P1: injectSkillCatalog now returns activeSkillIds (the filtered set that appears in the catalog). initializeAgent uses that set as the stored accessibleSkillIds on the initialized agent, so getSkillByName at runtime cannot resolve a deactivated skill — even if the LLM hallucinates a name or the user invokes by direct-invocation shorthand. Previously the executor authorized against the full ACL set, bypassing the active-state guarantee (Codex P1). P2: pruneOrphans now checks user access via findAccessibleResources in addition to skill existence. When a share is revoked, the user's skillStates entry for that skill had no cleanup path and silently consumed the 200-cap. Self-heals on both GET and POST. One extra ACL query per settings read/write; scoped to a single user so no N-user amplification (Codex P2). P2: the write queue moves from a single module-scoped object to a Map keyed by userId. Logout/login in the same tab can no longer flush the previous user's pending snapshot under the new session's auth. Each userId gets its own pending/inFlight slot; the in-flight request retains its original auth via the cookie already attached when sent, so the race window closes (Codex P2).
Address the remaining valid findings from the comprehensive review: - Extract toRecord, loadSkillStates, validateSkillStatesPayload, and pruneOrphanSkillStates into packages/api/src/skills/skillStates.ts as TypeScript. The controller in /api shrinks to a ~90-line thin wrapper that builds live dependency adapters for Mongoose + the permission service (Review #2 DRY, #3 workspace boundary) - Replace the triplicated 12-line skillStates loading block in initialize.js, openai.js, and responses.js with a single call to loadSkillStates from @librechat/api. One helper, three sites - Swap console.error for the project logger in the controller (Review #7) - Remove the redundant INVALID_KEY_PATTERN regex: a valid ObjectId cannot contain . or $, so isValidObjectIdString already covers it (Review #11) - Parameterize the 200-cap error toast with {{0}} interpolation driven by the error response's `limit` field, so future changes to MAX_SKILL_STATES update the UI message automatically (Review #12) - Add 24 unit tests for the new skillStates helpers (toRecord, resolveDefaultActiveOnShare, loadSkillStates, validateSkillStates- Payload, pruneOrphanSkillStates) covering success paths, malformed input, cap boundaries, and parallel-query behavior (Review #4) - Add 10 tests for injectSkillCatalog pagination covering empty accessible set, missing listSkillsByAccess, single-page filter, owned-vs-shared defaults, explicit-override precedence, multi-page collection, MAX_CATALOG_PAGES safety cap, early termination on has_more=false, additional_instructions injection, and fail-closed without userId (Review #5) Total test count: 60 (was 26 on this surface).
packages/api/src/types/error.ts already exports a ValidationError (MongooseError extension). Re-exporting a different shape from skills/skillStates.ts through the skills barrel caused TS2308 in CI because the root index re-exports both. Rename to SkillStatesValidationError to keep the exports disjoint.
Address the followup review findings: - Add optional `accessibleSkillIds` param to loadSkillStates so the helper short-circuits to defaults when no skills are accessible. All three controllers drop the residual 7-line conditional wrapper in favor of a single destructured call (Review #2) - Remove the unreachable `typeof key !== 'string'` check from validateSkillStatesPayload: Object.entries always yields string keys per the JS spec (Review #3) - Replace the two `as unknown as` agent casts in the injectSkillCatalog tests with a `makeAgent()` factory typed directly as the function's parameter shape (Review #4) - Tighten the MAX_CATALOG_PAGES assertion from `toBeLessThanOrEqual(11)` to `toHaveBeenCalledTimes(10)` — the loop deterministically makes exactly 10 page fetches before hitting the cap (Review #1) - Rewrite the parallel-execution test for pruneOrphanSkillStates using deferred promises instead of microtask-order assertions. The test now inspects `toHaveBeenCalledTimes(1)` on both mocks after a single Promise.resolve() yield, pinning Promise.all usage without relying on push-order into a shared array (Review #5) - Evict stale writeQueue entries on user change via a module-scoped `lastSeenUserId` sentinel. When a different user's toggle is the first one after a logout/login, the previous user's queue entry is deleted. Keeps the Map bounded without adding hook-instance effect cleanup (Review #6)
Collaborator
Author
|
@codex review |
The prior refactor replaced the inline 12-line skillStates loading block with a call to loadSkillStates from @librechat/api. Both controller spec files mock @librechat/api as a flat object, so any new named import from that package is undefined in the test env. Calling `await loadSkillStates(...)` threw before recordCollectedUsage ran, surfacing as "undefined is not iterable" on the test's array destructure of `mockRecordCollectedUsage.mock.calls[0]`. Add the missing mock to both spec files alongside the existing scopeSkillIds stub.
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 75eae4df61
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
Comment on lines
+107
to
+113
| const queue = getWriteQueue(userId); | ||
| while (queue.pending !== null) { | ||
| const next = queue.pending; | ||
| queue.pending = null; | ||
| queue.inFlight = true; | ||
| try { | ||
| await updateMutation.mutateAsync(next); |
There was a problem hiding this comment.
flush drains queue.pending without re-validating the active identity, so an in-flight loop can keep calling mutateAsync for a queue created under a previous userId. If a user toggles multiple times (creating pending state), then logs out and another user logs in before the first request settles, the stale queued snapshot can still be posted in the new session, overwriting the new user’s skillStates. Removing the old map entry does not prevent this because the loop retains a live reference to the old queue object.
Useful? React with 👍 / 👎.
Close the cross-session leak window where an in-flight flush loop still holds a reference to a previous user's queue: it could fire its next mutateAsync under the new session's auth cookies and persist the stale snapshot to the new user's document (Codex P1). Add an `abandoned` flag on `WriteQueue`. Three mechanisms cooperate: - `getWriteQueue` marks every non-active queue abandoned when the user differs from the last-seen identity (pre-existing eviction site, now more aggressive). - A `useEffect` on `userId` calls the same abandonment pass on every render with a new active identity, covering the window between logout/login and the new user's first toggle (when `getWriteQueue` would otherwise not fire). - The flush loop checks `!queue.abandoned` in its while condition so the second and later iterations exit without firing another `mutateAsync` after the session changes. The first iteration's in-flight request (already dispatched under the original user's cookies) still runs to completion or failure on its own — only the subsequent iterations, which are the dangerous ones, are blocked.
danny-avila
added a commit
that referenced
this pull request
Apr 17, 2026
…efaults (#12692) * feat: per-user skill active/inactive toggle with ownership-aware defaults - Add `skillStates` map (Record<string, boolean>) to user schema for per-user active/inactive overrides on skills - Add `defaultActiveOnShare` to interface.skills config (default: false) so admins can control whether shared skills auto-activate - Add GET/POST /api/user/settings/skills/active endpoints with validation - Add React Query hooks with optimistic mutations for skill states - Add useSkillActiveState hook with ownership-aware resolution: owned skills default active, shared skills default inactive - Add toggle switch UI to SkillListItem and SkillDetail components - Filter inactive skills in injectSkillCatalog before agent injection - Add localization keys for active/inactive labels * fix: use Record instead of Map for IUser.skillStates Mongoose .lean() flattens Map to a plain object, causing type incompatibility with IUser in methods that return lean documents. * fix: address review findings for skill active states - Fail-closed when userId is absent: filter rejects all shared skills instead of passing them through unfiltered (Codex P1) - Validate Mongoose Map key characters (reject . and $) in controller to return 400 instead of a 500 from schema validation (Codex P2) - Block toggle while initial skill states query is loading to prevent overwriting server-side overrides with an empty snapshot (Codex P2) - Extract shared SkillToggle component, eliminating duplicate toggle markup in SkillListItem and SkillDetail (Finding #3) - Move skill state query/mutation hooks from Favorites.ts to Skills/queries.ts per feature-directory convention (Finding #4) - Fix hardcoded English aria-label in SkillListItem by passing the localized string from the parent SkillList (Finding #5) - Fix inline arrow in SkillList render loop: pass stable callback reference so SkillListItem memo() is not invalidated (Finding #1) - Extract toRecord() helper in controller to DRY the Map-to-Object conversion (Finding #6) - Remove Promise.resolve wrapping synchronous config read (Finding #8) - Remove unused TUpdateSkillStatesRequest type (Finding #12) * fix: forward tabIndex on SkillToggle to preserve list keyboard nav The original inline toggle had tabIndex={-1} so the row itself remained the sole tab target. The extraction into SkillToggle dropped this prop, making every list toggle a tab stop. Add an optional tabIndex prop and pass -1 from SkillListItem. * fix: plumb skillStates to all agent entry points, isolate toggle keydown - Add skillStates/defaultActiveOnShare loading to openai.js and responses.js controllers so shared-skill activation is respected across all agent entry points, not just initialize.js (Codex P1) - Stop keydown propagation on SkillToggle so Enter/Space does not bubble to the parent row's navigation handler (Codex P2) * fix: paginate catalog fetch and serialize toggle writes - Paginate listSkillsByAccess (up to 10 pages of 100) until the active catalog quota is filled, so inactive shared skills in recent positions do not starve active owned skills past the first page (Codex P1) - Extend listSkillsByAccess interface with cursor/has_more/after for catalog pagination - Serialize skill-state writes via a ref queue: one in-flight request at a time, with the latest desired state sent when the previous one settles. Prevents last-response-wins races where an older request overwrites newer toggles (Codex P2) * fix: share write queue across hook instances, block toggle on fetch error - Move the write queue from a per-instance useRef to a module-scoped object so every mount of useSkillActiveState (SkillList, SkillDetail, etc.) serializes against the same in-flight slot. Prior per-instance queues allowed two components to race full-map POSTs (Codex P1) - Extend the toggle guard beyond isLoading: also block when isError is true or data is undefined. Prevents a failed GET from seeding a toggle with an empty baseline that would wipe server-side overrides on the next successful POST (Codex P1) * fix: stale closure, orphan cleanup, and cap-error UX - Read toggle baseline from React Query cache via queryClient.getQueryData instead of the captured skillStates closure. The closure can be stale between onMutate's setQueryData and the next render, so rapid successive toggles would build on old state and drop earlier changes (Codex P1) - Surface the MAX_SKILL_STATES_EXCEEDED error code with a specific toast key (com_ui_skill_states_limit) so users understand the 200-cap rather than seeing a generic error - Prune orphaned entries (skillIds whose Skill doc no longer exists) on both GET and POST in SkillStatesController. Self-heals over time without needing cascade-delete hooks or a migration job. Uses one indexed Skill._id query per request * test: pin skill active-state precedence with unit tests Extract the active-state resolution logic from a closure inside injectSkillCatalog into an exported resolveSkillActive helper, then cover every branch of the precedence matrix: - Fails closed when userId is absent (even with defaultActiveOnShare=true) - Explicit override wins over ownership and config (both true and false) - Owned skills default to active when no override is set - Shared skills default to defaultActiveOnShare value - Undefined skillStates behaves identically to an empty object - defaultActiveOnShare defaults to false when omitted - Owned skills ignore defaultActiveOnShare entirely Closes Finding #2 from the pre-rebase comprehensive review. Mirrors the existing scopeSkillIds test style; injectSkillCatalog now calls resolveSkillActive instead of inlining the closure. * refactor: limit skill active toggle to detail header, drop label - Remove the per-row toggle from SkillListItem and the active-state plumbing (hook call, isSkillEnabled/onToggleEnabled/toggleAriaLabel props) from SkillList. The detail view is now the single place to change a skill's active state - Drop dim/muted styling for inactive skills in the sidebar: without a control there, the visual indication has nowhere to land - Resize SkillToggle to match neighbor buttons: outer h-9 container, h-6 w-11 track with size-5 knob, no label span. The 'Active' / 'Inactive' text that accompanied the detail-view toggle is removed - Remove the now-unused label prop and tabIndex prop (the tabIndex existed only for the list-row context) from SkillToggle. Drop the onKeyDown stopPropagation for the same reason - Remove now-orphaned com_ui_skill_active / com_ui_skill_inactive translation keys * style: shrink SkillToggle track to h-5 w-9 with size-4 knob Container stays at h-9 to match neighbor button heights. The toggle track itself drops from h-6 w-11 to h-5 w-9, with a size-4 knob travelling 1.125rem on activation. Visually lighter inside the row. * fix: remove redundant skillStates entries that match the resolved default When a toggle lands on the ownership/config default, delete the key from the map instead of persisting `{id: defaultValue}`. Without this, a user toggling a skill off and back on would leave `{id: true}` for an owned skill (whose default is already true), silently consuming a slot against the 200-entry cap. Repeated round-trip toggles could exhaust the quota with zero meaningful overrides (Codex P2). Preserves the exceptions-list invariant that the runtime-resolution design depends on. * fix: prune before enforcing skill-state cap; reject non-ObjectId keys Reorder the update controller so pruneOrphans runs before the 200-cap check. Without this, a user near the cap with some orphaned entries (skills deleted since their last GET) could send a payload that would pass after pruning but gets rejected by the raw-size check first. Add a sanity cap on raw payload size (2 * MAX_SKILL_STATES) so abusive inputs do not reach the DB query, and enforce the real cap on the pruned result instead. Harden pruneOrphans: the earlier early-return path could pass non-ObjectId keys through unchanged. Now only valid ObjectIds are returned, and the Skill-model-unavailable fallback filters by format. Also add isValidObjectIdString validation at the input boundary so malformed (but otherwise non-Mongo-unsafe) keys never reach persistence (Codex P2 x2). * fix: enforce active filter at execute time, prune revoked shares, scope queue per user P1: injectSkillCatalog now returns activeSkillIds (the filtered set that appears in the catalog). initializeAgent uses that set as the stored accessibleSkillIds on the initialized agent, so getSkillByName at runtime cannot resolve a deactivated skill — even if the LLM hallucinates a name or the user invokes by direct-invocation shorthand. Previously the executor authorized against the full ACL set, bypassing the active-state guarantee (Codex P1). P2: pruneOrphans now checks user access via findAccessibleResources in addition to skill existence. When a share is revoked, the user's skillStates entry for that skill had no cleanup path and silently consumed the 200-cap. Self-heals on both GET and POST. One extra ACL query per settings read/write; scoped to a single user so no N-user amplification (Codex P2). P2: the write queue moves from a single module-scoped object to a Map keyed by userId. Logout/login in the same tab can no longer flush the previous user's pending snapshot under the new session's auth. Each userId gets its own pending/inFlight slot; the in-flight request retains its original auth via the cookie already attached when sent, so the race window closes (Codex P2). * refactor: extract skillStates helpers to packages/api; add tests; polish Address the remaining valid findings from the comprehensive review: - Extract toRecord, loadSkillStates, validateSkillStatesPayload, and pruneOrphanSkillStates into packages/api/src/skills/skillStates.ts as TypeScript. The controller in /api shrinks to a ~90-line thin wrapper that builds live dependency adapters for Mongoose + the permission service (Review #2 DRY, #3 workspace boundary) - Replace the triplicated 12-line skillStates loading block in initialize.js, openai.js, and responses.js with a single call to loadSkillStates from @librechat/api. One helper, three sites - Swap console.error for the project logger in the controller (Review #7) - Remove the redundant INVALID_KEY_PATTERN regex: a valid ObjectId cannot contain . or $, so isValidObjectIdString already covers it (Review #11) - Parameterize the 200-cap error toast with {{0}} interpolation driven by the error response's `limit` field, so future changes to MAX_SKILL_STATES update the UI message automatically (Review #12) - Add 24 unit tests for the new skillStates helpers (toRecord, resolveDefaultActiveOnShare, loadSkillStates, validateSkillStates- Payload, pruneOrphanSkillStates) covering success paths, malformed input, cap boundaries, and parallel-query behavior (Review #4) - Add 10 tests for injectSkillCatalog pagination covering empty accessible set, missing listSkillsByAccess, single-page filter, owned-vs-shared defaults, explicit-override precedence, multi-page collection, MAX_CATALOG_PAGES safety cap, early termination on has_more=false, additional_instructions injection, and fail-closed without userId (Review #5) Total test count: 60 (was 26 on this surface). * fix: rename skillStates ValidationError to avoid barrel-export collision packages/api/src/types/error.ts already exports a ValidationError (MongooseError extension). Re-exporting a different shape from skills/skillStates.ts through the skills barrel caused TS2308 in CI because the root index re-exports both. Rename to SkillStatesValidationError to keep the exports disjoint. * refactor: tighten tests and absorb caller guard into loadSkillStates Address the followup review findings: - Add optional `accessibleSkillIds` param to loadSkillStates so the helper short-circuits to defaults when no skills are accessible. All three controllers drop the residual 7-line conditional wrapper in favor of a single destructured call (Review #2) - Remove the unreachable `typeof key !== 'string'` check from validateSkillStatesPayload: Object.entries always yields string keys per the JS spec (Review #3) - Replace the two `as unknown as` agent casts in the injectSkillCatalog tests with a `makeAgent()` factory typed directly as the function's parameter shape (Review #4) - Tighten the MAX_CATALOG_PAGES assertion from `toBeLessThanOrEqual(11)` to `toHaveBeenCalledTimes(10)` — the loop deterministically makes exactly 10 page fetches before hitting the cap (Review #1) - Rewrite the parallel-execution test for pruneOrphanSkillStates using deferred promises instead of microtask-order assertions. The test now inspects `toHaveBeenCalledTimes(1)` on both mocks after a single Promise.resolve() yield, pinning Promise.all usage without relying on push-order into a shared array (Review #5) - Evict stale writeQueue entries on user change via a module-scoped `lastSeenUserId` sentinel. When a different user's toggle is the first one after a logout/login, the previous user's queue entry is deleted. Keeps the Map bounded without adding hook-instance effect cleanup (Review #6) * fix(test): mock loadSkillStates in openai and responses controller specs The prior refactor replaced the inline 12-line skillStates loading block with a call to loadSkillStates from @librechat/api. Both controller spec files mock @librechat/api as a flat object, so any new named import from that package is undefined in the test env. Calling `await loadSkillStates(...)` threw before recordCollectedUsage ran, surfacing as "undefined is not iterable" on the test's array destructure of `mockRecordCollectedUsage.mock.calls[0]`. Add the missing mock to both spec files alongside the existing scopeSkillIds stub. * fix: abandon stale skillStates write queues on user switch Close the cross-session leak window where an in-flight flush loop still holds a reference to a previous user's queue: it could fire its next mutateAsync under the new session's auth cookies and persist the stale snapshot to the new user's document (Codex P1). Add an `abandoned` flag on `WriteQueue`. Three mechanisms cooperate: - `getWriteQueue` marks every non-active queue abandoned when the user differs from the last-seen identity (pre-existing eviction site, now more aggressive). - A `useEffect` on `userId` calls the same abandonment pass on every render with a new active identity, covering the window between logout/login and the new user's first toggle (when `getWriteQueue` would otherwise not fire). - The flush loop checks `!queue.abandoned` in its while condition so the second and later iterations exit without firing another `mutateAsync` after the session changes. The first iteration's in-flight request (already dispatched under the original user's cookies) still runs to completion or failure on its own — only the subsequent iterations, which are the dangerous ones, are blocked.
danny-avila
added a commit
that referenced
this pull request
Apr 21, 2026
…efaults (#12692) * feat: per-user skill active/inactive toggle with ownership-aware defaults - Add `skillStates` map (Record<string, boolean>) to user schema for per-user active/inactive overrides on skills - Add `defaultActiveOnShare` to interface.skills config (default: false) so admins can control whether shared skills auto-activate - Add GET/POST /api/user/settings/skills/active endpoints with validation - Add React Query hooks with optimistic mutations for skill states - Add useSkillActiveState hook with ownership-aware resolution: owned skills default active, shared skills default inactive - Add toggle switch UI to SkillListItem and SkillDetail components - Filter inactive skills in injectSkillCatalog before agent injection - Add localization keys for active/inactive labels * fix: use Record instead of Map for IUser.skillStates Mongoose .lean() flattens Map to a plain object, causing type incompatibility with IUser in methods that return lean documents. * fix: address review findings for skill active states - Fail-closed when userId is absent: filter rejects all shared skills instead of passing them through unfiltered (Codex P1) - Validate Mongoose Map key characters (reject . and $) in controller to return 400 instead of a 500 from schema validation (Codex P2) - Block toggle while initial skill states query is loading to prevent overwriting server-side overrides with an empty snapshot (Codex P2) - Extract shared SkillToggle component, eliminating duplicate toggle markup in SkillListItem and SkillDetail (Finding #3) - Move skill state query/mutation hooks from Favorites.ts to Skills/queries.ts per feature-directory convention (Finding #4) - Fix hardcoded English aria-label in SkillListItem by passing the localized string from the parent SkillList (Finding #5) - Fix inline arrow in SkillList render loop: pass stable callback reference so SkillListItem memo() is not invalidated (Finding #1) - Extract toRecord() helper in controller to DRY the Map-to-Object conversion (Finding #6) - Remove Promise.resolve wrapping synchronous config read (Finding #8) - Remove unused TUpdateSkillStatesRequest type (Finding #12) * fix: forward tabIndex on SkillToggle to preserve list keyboard nav The original inline toggle had tabIndex={-1} so the row itself remained the sole tab target. The extraction into SkillToggle dropped this prop, making every list toggle a tab stop. Add an optional tabIndex prop and pass -1 from SkillListItem. * fix: plumb skillStates to all agent entry points, isolate toggle keydown - Add skillStates/defaultActiveOnShare loading to openai.js and responses.js controllers so shared-skill activation is respected across all agent entry points, not just initialize.js (Codex P1) - Stop keydown propagation on SkillToggle so Enter/Space does not bubble to the parent row's navigation handler (Codex P2) * fix: paginate catalog fetch and serialize toggle writes - Paginate listSkillsByAccess (up to 10 pages of 100) until the active catalog quota is filled, so inactive shared skills in recent positions do not starve active owned skills past the first page (Codex P1) - Extend listSkillsByAccess interface with cursor/has_more/after for catalog pagination - Serialize skill-state writes via a ref queue: one in-flight request at a time, with the latest desired state sent when the previous one settles. Prevents last-response-wins races where an older request overwrites newer toggles (Codex P2) * fix: share write queue across hook instances, block toggle on fetch error - Move the write queue from a per-instance useRef to a module-scoped object so every mount of useSkillActiveState (SkillList, SkillDetail, etc.) serializes against the same in-flight slot. Prior per-instance queues allowed two components to race full-map POSTs (Codex P1) - Extend the toggle guard beyond isLoading: also block when isError is true or data is undefined. Prevents a failed GET from seeding a toggle with an empty baseline that would wipe server-side overrides on the next successful POST (Codex P1) * fix: stale closure, orphan cleanup, and cap-error UX - Read toggle baseline from React Query cache via queryClient.getQueryData instead of the captured skillStates closure. The closure can be stale between onMutate's setQueryData and the next render, so rapid successive toggles would build on old state and drop earlier changes (Codex P1) - Surface the MAX_SKILL_STATES_EXCEEDED error code with a specific toast key (com_ui_skill_states_limit) so users understand the 200-cap rather than seeing a generic error - Prune orphaned entries (skillIds whose Skill doc no longer exists) on both GET and POST in SkillStatesController. Self-heals over time without needing cascade-delete hooks or a migration job. Uses one indexed Skill._id query per request * test: pin skill active-state precedence with unit tests Extract the active-state resolution logic from a closure inside injectSkillCatalog into an exported resolveSkillActive helper, then cover every branch of the precedence matrix: - Fails closed when userId is absent (even with defaultActiveOnShare=true) - Explicit override wins over ownership and config (both true and false) - Owned skills default to active when no override is set - Shared skills default to defaultActiveOnShare value - Undefined skillStates behaves identically to an empty object - defaultActiveOnShare defaults to false when omitted - Owned skills ignore defaultActiveOnShare entirely Closes Finding #2 from the pre-rebase comprehensive review. Mirrors the existing scopeSkillIds test style; injectSkillCatalog now calls resolveSkillActive instead of inlining the closure. * refactor: limit skill active toggle to detail header, drop label - Remove the per-row toggle from SkillListItem and the active-state plumbing (hook call, isSkillEnabled/onToggleEnabled/toggleAriaLabel props) from SkillList. The detail view is now the single place to change a skill's active state - Drop dim/muted styling for inactive skills in the sidebar: without a control there, the visual indication has nowhere to land - Resize SkillToggle to match neighbor buttons: outer h-9 container, h-6 w-11 track with size-5 knob, no label span. The 'Active' / 'Inactive' text that accompanied the detail-view toggle is removed - Remove the now-unused label prop and tabIndex prop (the tabIndex existed only for the list-row context) from SkillToggle. Drop the onKeyDown stopPropagation for the same reason - Remove now-orphaned com_ui_skill_active / com_ui_skill_inactive translation keys * style: shrink SkillToggle track to h-5 w-9 with size-4 knob Container stays at h-9 to match neighbor button heights. The toggle track itself drops from h-6 w-11 to h-5 w-9, with a size-4 knob travelling 1.125rem on activation. Visually lighter inside the row. * fix: remove redundant skillStates entries that match the resolved default When a toggle lands on the ownership/config default, delete the key from the map instead of persisting `{id: defaultValue}`. Without this, a user toggling a skill off and back on would leave `{id: true}` for an owned skill (whose default is already true), silently consuming a slot against the 200-entry cap. Repeated round-trip toggles could exhaust the quota with zero meaningful overrides (Codex P2). Preserves the exceptions-list invariant that the runtime-resolution design depends on. * fix: prune before enforcing skill-state cap; reject non-ObjectId keys Reorder the update controller so pruneOrphans runs before the 200-cap check. Without this, a user near the cap with some orphaned entries (skills deleted since their last GET) could send a payload that would pass after pruning but gets rejected by the raw-size check first. Add a sanity cap on raw payload size (2 * MAX_SKILL_STATES) so abusive inputs do not reach the DB query, and enforce the real cap on the pruned result instead. Harden pruneOrphans: the earlier early-return path could pass non-ObjectId keys through unchanged. Now only valid ObjectIds are returned, and the Skill-model-unavailable fallback filters by format. Also add isValidObjectIdString validation at the input boundary so malformed (but otherwise non-Mongo-unsafe) keys never reach persistence (Codex P2 x2). * fix: enforce active filter at execute time, prune revoked shares, scope queue per user P1: injectSkillCatalog now returns activeSkillIds (the filtered set that appears in the catalog). initializeAgent uses that set as the stored accessibleSkillIds on the initialized agent, so getSkillByName at runtime cannot resolve a deactivated skill — even if the LLM hallucinates a name or the user invokes by direct-invocation shorthand. Previously the executor authorized against the full ACL set, bypassing the active-state guarantee (Codex P1). P2: pruneOrphans now checks user access via findAccessibleResources in addition to skill existence. When a share is revoked, the user's skillStates entry for that skill had no cleanup path and silently consumed the 200-cap. Self-heals on both GET and POST. One extra ACL query per settings read/write; scoped to a single user so no N-user amplification (Codex P2). P2: the write queue moves from a single module-scoped object to a Map keyed by userId. Logout/login in the same tab can no longer flush the previous user's pending snapshot under the new session's auth. Each userId gets its own pending/inFlight slot; the in-flight request retains its original auth via the cookie already attached when sent, so the race window closes (Codex P2). * refactor: extract skillStates helpers to packages/api; add tests; polish Address the remaining valid findings from the comprehensive review: - Extract toRecord, loadSkillStates, validateSkillStatesPayload, and pruneOrphanSkillStates into packages/api/src/skills/skillStates.ts as TypeScript. The controller in /api shrinks to a ~90-line thin wrapper that builds live dependency adapters for Mongoose + the permission service (Review #2 DRY, #3 workspace boundary) - Replace the triplicated 12-line skillStates loading block in initialize.js, openai.js, and responses.js with a single call to loadSkillStates from @librechat/api. One helper, three sites - Swap console.error for the project logger in the controller (Review #7) - Remove the redundant INVALID_KEY_PATTERN regex: a valid ObjectId cannot contain . or $, so isValidObjectIdString already covers it (Review #11) - Parameterize the 200-cap error toast with {{0}} interpolation driven by the error response's `limit` field, so future changes to MAX_SKILL_STATES update the UI message automatically (Review #12) - Add 24 unit tests for the new skillStates helpers (toRecord, resolveDefaultActiveOnShare, loadSkillStates, validateSkillStates- Payload, pruneOrphanSkillStates) covering success paths, malformed input, cap boundaries, and parallel-query behavior (Review #4) - Add 10 tests for injectSkillCatalog pagination covering empty accessible set, missing listSkillsByAccess, single-page filter, owned-vs-shared defaults, explicit-override precedence, multi-page collection, MAX_CATALOG_PAGES safety cap, early termination on has_more=false, additional_instructions injection, and fail-closed without userId (Review #5) Total test count: 60 (was 26 on this surface). * fix: rename skillStates ValidationError to avoid barrel-export collision packages/api/src/types/error.ts already exports a ValidationError (MongooseError extension). Re-exporting a different shape from skills/skillStates.ts through the skills barrel caused TS2308 in CI because the root index re-exports both. Rename to SkillStatesValidationError to keep the exports disjoint. * refactor: tighten tests and absorb caller guard into loadSkillStates Address the followup review findings: - Add optional `accessibleSkillIds` param to loadSkillStates so the helper short-circuits to defaults when no skills are accessible. All three controllers drop the residual 7-line conditional wrapper in favor of a single destructured call (Review #2) - Remove the unreachable `typeof key !== 'string'` check from validateSkillStatesPayload: Object.entries always yields string keys per the JS spec (Review #3) - Replace the two `as unknown as` agent casts in the injectSkillCatalog tests with a `makeAgent()` factory typed directly as the function's parameter shape (Review #4) - Tighten the MAX_CATALOG_PAGES assertion from `toBeLessThanOrEqual(11)` to `toHaveBeenCalledTimes(10)` — the loop deterministically makes exactly 10 page fetches before hitting the cap (Review #1) - Rewrite the parallel-execution test for pruneOrphanSkillStates using deferred promises instead of microtask-order assertions. The test now inspects `toHaveBeenCalledTimes(1)` on both mocks after a single Promise.resolve() yield, pinning Promise.all usage without relying on push-order into a shared array (Review #5) - Evict stale writeQueue entries on user change via a module-scoped `lastSeenUserId` sentinel. When a different user's toggle is the first one after a logout/login, the previous user's queue entry is deleted. Keeps the Map bounded without adding hook-instance effect cleanup (Review #6) * fix(test): mock loadSkillStates in openai and responses controller specs The prior refactor replaced the inline 12-line skillStates loading block with a call to loadSkillStates from @librechat/api. Both controller spec files mock @librechat/api as a flat object, so any new named import from that package is undefined in the test env. Calling `await loadSkillStates(...)` threw before recordCollectedUsage ran, surfacing as "undefined is not iterable" on the test's array destructure of `mockRecordCollectedUsage.mock.calls[0]`. Add the missing mock to both spec files alongside the existing scopeSkillIds stub. * fix: abandon stale skillStates write queues on user switch Close the cross-session leak window where an in-flight flush loop still holds a reference to a previous user's queue: it could fire its next mutateAsync under the new session's auth cookies and persist the stale snapshot to the new user's document (Codex P1). Add an `abandoned` flag on `WriteQueue`. Three mechanisms cooperate: - `getWriteQueue` marks every non-active queue abandoned when the user differs from the last-seen identity (pre-existing eviction site, now more aggressive). - A `useEffect` on `userId` calls the same abandonment pass on every render with a new active identity, covering the window between logout/login and the new user's first toggle (when `getWriteQueue` would otherwise not fire). - The flush loop checks `!queue.abandoned` in its while condition so the second and later iterations exit without firing another `mutateAsync` after the session changes. The first iteration's in-flight request (already dispatched under the original user's cookies) still runs to completion or failure on its own — only the subsequent iterations, which are the dangerous ones, are blocked.
danny-avila
added a commit
that referenced
this pull request
Apr 22, 2026
…efaults (#12692) * feat: per-user skill active/inactive toggle with ownership-aware defaults - Add `skillStates` map (Record<string, boolean>) to user schema for per-user active/inactive overrides on skills - Add `defaultActiveOnShare` to interface.skills config (default: false) so admins can control whether shared skills auto-activate - Add GET/POST /api/user/settings/skills/active endpoints with validation - Add React Query hooks with optimistic mutations for skill states - Add useSkillActiveState hook with ownership-aware resolution: owned skills default active, shared skills default inactive - Add toggle switch UI to SkillListItem and SkillDetail components - Filter inactive skills in injectSkillCatalog before agent injection - Add localization keys for active/inactive labels * fix: use Record instead of Map for IUser.skillStates Mongoose .lean() flattens Map to a plain object, causing type incompatibility with IUser in methods that return lean documents. * fix: address review findings for skill active states - Fail-closed when userId is absent: filter rejects all shared skills instead of passing them through unfiltered (Codex P1) - Validate Mongoose Map key characters (reject . and $) in controller to return 400 instead of a 500 from schema validation (Codex P2) - Block toggle while initial skill states query is loading to prevent overwriting server-side overrides with an empty snapshot (Codex P2) - Extract shared SkillToggle component, eliminating duplicate toggle markup in SkillListItem and SkillDetail (Finding #3) - Move skill state query/mutation hooks from Favorites.ts to Skills/queries.ts per feature-directory convention (Finding #4) - Fix hardcoded English aria-label in SkillListItem by passing the localized string from the parent SkillList (Finding #5) - Fix inline arrow in SkillList render loop: pass stable callback reference so SkillListItem memo() is not invalidated (Finding #1) - Extract toRecord() helper in controller to DRY the Map-to-Object conversion (Finding #6) - Remove Promise.resolve wrapping synchronous config read (Finding #8) - Remove unused TUpdateSkillStatesRequest type (Finding #12) * fix: forward tabIndex on SkillToggle to preserve list keyboard nav The original inline toggle had tabIndex={-1} so the row itself remained the sole tab target. The extraction into SkillToggle dropped this prop, making every list toggle a tab stop. Add an optional tabIndex prop and pass -1 from SkillListItem. * fix: plumb skillStates to all agent entry points, isolate toggle keydown - Add skillStates/defaultActiveOnShare loading to openai.js and responses.js controllers so shared-skill activation is respected across all agent entry points, not just initialize.js (Codex P1) - Stop keydown propagation on SkillToggle so Enter/Space does not bubble to the parent row's navigation handler (Codex P2) * fix: paginate catalog fetch and serialize toggle writes - Paginate listSkillsByAccess (up to 10 pages of 100) until the active catalog quota is filled, so inactive shared skills in recent positions do not starve active owned skills past the first page (Codex P1) - Extend listSkillsByAccess interface with cursor/has_more/after for catalog pagination - Serialize skill-state writes via a ref queue: one in-flight request at a time, with the latest desired state sent when the previous one settles. Prevents last-response-wins races where an older request overwrites newer toggles (Codex P2) * fix: share write queue across hook instances, block toggle on fetch error - Move the write queue from a per-instance useRef to a module-scoped object so every mount of useSkillActiveState (SkillList, SkillDetail, etc.) serializes against the same in-flight slot. Prior per-instance queues allowed two components to race full-map POSTs (Codex P1) - Extend the toggle guard beyond isLoading: also block when isError is true or data is undefined. Prevents a failed GET from seeding a toggle with an empty baseline that would wipe server-side overrides on the next successful POST (Codex P1) * fix: stale closure, orphan cleanup, and cap-error UX - Read toggle baseline from React Query cache via queryClient.getQueryData instead of the captured skillStates closure. The closure can be stale between onMutate's setQueryData and the next render, so rapid successive toggles would build on old state and drop earlier changes (Codex P1) - Surface the MAX_SKILL_STATES_EXCEEDED error code with a specific toast key (com_ui_skill_states_limit) so users understand the 200-cap rather than seeing a generic error - Prune orphaned entries (skillIds whose Skill doc no longer exists) on both GET and POST in SkillStatesController. Self-heals over time without needing cascade-delete hooks or a migration job. Uses one indexed Skill._id query per request * test: pin skill active-state precedence with unit tests Extract the active-state resolution logic from a closure inside injectSkillCatalog into an exported resolveSkillActive helper, then cover every branch of the precedence matrix: - Fails closed when userId is absent (even with defaultActiveOnShare=true) - Explicit override wins over ownership and config (both true and false) - Owned skills default to active when no override is set - Shared skills default to defaultActiveOnShare value - Undefined skillStates behaves identically to an empty object - defaultActiveOnShare defaults to false when omitted - Owned skills ignore defaultActiveOnShare entirely Closes Finding #2 from the pre-rebase comprehensive review. Mirrors the existing scopeSkillIds test style; injectSkillCatalog now calls resolveSkillActive instead of inlining the closure. * refactor: limit skill active toggle to detail header, drop label - Remove the per-row toggle from SkillListItem and the active-state plumbing (hook call, isSkillEnabled/onToggleEnabled/toggleAriaLabel props) from SkillList. The detail view is now the single place to change a skill's active state - Drop dim/muted styling for inactive skills in the sidebar: without a control there, the visual indication has nowhere to land - Resize SkillToggle to match neighbor buttons: outer h-9 container, h-6 w-11 track with size-5 knob, no label span. The 'Active' / 'Inactive' text that accompanied the detail-view toggle is removed - Remove the now-unused label prop and tabIndex prop (the tabIndex existed only for the list-row context) from SkillToggle. Drop the onKeyDown stopPropagation for the same reason - Remove now-orphaned com_ui_skill_active / com_ui_skill_inactive translation keys * style: shrink SkillToggle track to h-5 w-9 with size-4 knob Container stays at h-9 to match neighbor button heights. The toggle track itself drops from h-6 w-11 to h-5 w-9, with a size-4 knob travelling 1.125rem on activation. Visually lighter inside the row. * fix: remove redundant skillStates entries that match the resolved default When a toggle lands on the ownership/config default, delete the key from the map instead of persisting `{id: defaultValue}`. Without this, a user toggling a skill off and back on would leave `{id: true}` for an owned skill (whose default is already true), silently consuming a slot against the 200-entry cap. Repeated round-trip toggles could exhaust the quota with zero meaningful overrides (Codex P2). Preserves the exceptions-list invariant that the runtime-resolution design depends on. * fix: prune before enforcing skill-state cap; reject non-ObjectId keys Reorder the update controller so pruneOrphans runs before the 200-cap check. Without this, a user near the cap with some orphaned entries (skills deleted since their last GET) could send a payload that would pass after pruning but gets rejected by the raw-size check first. Add a sanity cap on raw payload size (2 * MAX_SKILL_STATES) so abusive inputs do not reach the DB query, and enforce the real cap on the pruned result instead. Harden pruneOrphans: the earlier early-return path could pass non-ObjectId keys through unchanged. Now only valid ObjectIds are returned, and the Skill-model-unavailable fallback filters by format. Also add isValidObjectIdString validation at the input boundary so malformed (but otherwise non-Mongo-unsafe) keys never reach persistence (Codex P2 x2). * fix: enforce active filter at execute time, prune revoked shares, scope queue per user P1: injectSkillCatalog now returns activeSkillIds (the filtered set that appears in the catalog). initializeAgent uses that set as the stored accessibleSkillIds on the initialized agent, so getSkillByName at runtime cannot resolve a deactivated skill — even if the LLM hallucinates a name or the user invokes by direct-invocation shorthand. Previously the executor authorized against the full ACL set, bypassing the active-state guarantee (Codex P1). P2: pruneOrphans now checks user access via findAccessibleResources in addition to skill existence. When a share is revoked, the user's skillStates entry for that skill had no cleanup path and silently consumed the 200-cap. Self-heals on both GET and POST. One extra ACL query per settings read/write; scoped to a single user so no N-user amplification (Codex P2). P2: the write queue moves from a single module-scoped object to a Map keyed by userId. Logout/login in the same tab can no longer flush the previous user's pending snapshot under the new session's auth. Each userId gets its own pending/inFlight slot; the in-flight request retains its original auth via the cookie already attached when sent, so the race window closes (Codex P2). * refactor: extract skillStates helpers to packages/api; add tests; polish Address the remaining valid findings from the comprehensive review: - Extract toRecord, loadSkillStates, validateSkillStatesPayload, and pruneOrphanSkillStates into packages/api/src/skills/skillStates.ts as TypeScript. The controller in /api shrinks to a ~90-line thin wrapper that builds live dependency adapters for Mongoose + the permission service (Review #2 DRY, #3 workspace boundary) - Replace the triplicated 12-line skillStates loading block in initialize.js, openai.js, and responses.js with a single call to loadSkillStates from @librechat/api. One helper, three sites - Swap console.error for the project logger in the controller (Review #7) - Remove the redundant INVALID_KEY_PATTERN regex: a valid ObjectId cannot contain . or $, so isValidObjectIdString already covers it (Review #11) - Parameterize the 200-cap error toast with {{0}} interpolation driven by the error response's `limit` field, so future changes to MAX_SKILL_STATES update the UI message automatically (Review #12) - Add 24 unit tests for the new skillStates helpers (toRecord, resolveDefaultActiveOnShare, loadSkillStates, validateSkillStates- Payload, pruneOrphanSkillStates) covering success paths, malformed input, cap boundaries, and parallel-query behavior (Review #4) - Add 10 tests for injectSkillCatalog pagination covering empty accessible set, missing listSkillsByAccess, single-page filter, owned-vs-shared defaults, explicit-override precedence, multi-page collection, MAX_CATALOG_PAGES safety cap, early termination on has_more=false, additional_instructions injection, and fail-closed without userId (Review #5) Total test count: 60 (was 26 on this surface). * fix: rename skillStates ValidationError to avoid barrel-export collision packages/api/src/types/error.ts already exports a ValidationError (MongooseError extension). Re-exporting a different shape from skills/skillStates.ts through the skills barrel caused TS2308 in CI because the root index re-exports both. Rename to SkillStatesValidationError to keep the exports disjoint. * refactor: tighten tests and absorb caller guard into loadSkillStates Address the followup review findings: - Add optional `accessibleSkillIds` param to loadSkillStates so the helper short-circuits to defaults when no skills are accessible. All three controllers drop the residual 7-line conditional wrapper in favor of a single destructured call (Review #2) - Remove the unreachable `typeof key !== 'string'` check from validateSkillStatesPayload: Object.entries always yields string keys per the JS spec (Review #3) - Replace the two `as unknown as` agent casts in the injectSkillCatalog tests with a `makeAgent()` factory typed directly as the function's parameter shape (Review #4) - Tighten the MAX_CATALOG_PAGES assertion from `toBeLessThanOrEqual(11)` to `toHaveBeenCalledTimes(10)` — the loop deterministically makes exactly 10 page fetches before hitting the cap (Review #1) - Rewrite the parallel-execution test for pruneOrphanSkillStates using deferred promises instead of microtask-order assertions. The test now inspects `toHaveBeenCalledTimes(1)` on both mocks after a single Promise.resolve() yield, pinning Promise.all usage without relying on push-order into a shared array (Review #5) - Evict stale writeQueue entries on user change via a module-scoped `lastSeenUserId` sentinel. When a different user's toggle is the first one after a logout/login, the previous user's queue entry is deleted. Keeps the Map bounded without adding hook-instance effect cleanup (Review #6) * fix(test): mock loadSkillStates in openai and responses controller specs The prior refactor replaced the inline 12-line skillStates loading block with a call to loadSkillStates from @librechat/api. Both controller spec files mock @librechat/api as a flat object, so any new named import from that package is undefined in the test env. Calling `await loadSkillStates(...)` threw before recordCollectedUsage ran, surfacing as "undefined is not iterable" on the test's array destructure of `mockRecordCollectedUsage.mock.calls[0]`. Add the missing mock to both spec files alongside the existing scopeSkillIds stub. * fix: abandon stale skillStates write queues on user switch Close the cross-session leak window where an in-flight flush loop still holds a reference to a previous user's queue: it could fire its next mutateAsync under the new session's auth cookies and persist the stale snapshot to the new user's document (Codex P1). Add an `abandoned` flag on `WriteQueue`. Three mechanisms cooperate: - `getWriteQueue` marks every non-active queue abandoned when the user differs from the last-seen identity (pre-existing eviction site, now more aggressive). - A `useEffect` on `userId` calls the same abandonment pass on every render with a new active identity, covering the window between logout/login and the new user's first toggle (when `getWriteQueue` would otherwise not fire). - The flush loop checks `!queue.abandoned` in its while condition so the second and later iterations exit without firing another `mutateAsync` after the session changes. The first iteration's in-flight request (already dispatched under the original user's cookies) still runs to completion or failure on its own — only the subsequent iterations, which are the dangerous ones, are blocked.
danny-avila
added a commit
that referenced
this pull request
Apr 25, 2026
…efaults (#12692) * feat: per-user skill active/inactive toggle with ownership-aware defaults - Add `skillStates` map (Record<string, boolean>) to user schema for per-user active/inactive overrides on skills - Add `defaultActiveOnShare` to interface.skills config (default: false) so admins can control whether shared skills auto-activate - Add GET/POST /api/user/settings/skills/active endpoints with validation - Add React Query hooks with optimistic mutations for skill states - Add useSkillActiveState hook with ownership-aware resolution: owned skills default active, shared skills default inactive - Add toggle switch UI to SkillListItem and SkillDetail components - Filter inactive skills in injectSkillCatalog before agent injection - Add localization keys for active/inactive labels * fix: use Record instead of Map for IUser.skillStates Mongoose .lean() flattens Map to a plain object, causing type incompatibility with IUser in methods that return lean documents. * fix: address review findings for skill active states - Fail-closed when userId is absent: filter rejects all shared skills instead of passing them through unfiltered (Codex P1) - Validate Mongoose Map key characters (reject . and $) in controller to return 400 instead of a 500 from schema validation (Codex P2) - Block toggle while initial skill states query is loading to prevent overwriting server-side overrides with an empty snapshot (Codex P2) - Extract shared SkillToggle component, eliminating duplicate toggle markup in SkillListItem and SkillDetail (Finding #3) - Move skill state query/mutation hooks from Favorites.ts to Skills/queries.ts per feature-directory convention (Finding #4) - Fix hardcoded English aria-label in SkillListItem by passing the localized string from the parent SkillList (Finding #5) - Fix inline arrow in SkillList render loop: pass stable callback reference so SkillListItem memo() is not invalidated (Finding #1) - Extract toRecord() helper in controller to DRY the Map-to-Object conversion (Finding #6) - Remove Promise.resolve wrapping synchronous config read (Finding #8) - Remove unused TUpdateSkillStatesRequest type (Finding #12) * fix: forward tabIndex on SkillToggle to preserve list keyboard nav The original inline toggle had tabIndex={-1} so the row itself remained the sole tab target. The extraction into SkillToggle dropped this prop, making every list toggle a tab stop. Add an optional tabIndex prop and pass -1 from SkillListItem. * fix: plumb skillStates to all agent entry points, isolate toggle keydown - Add skillStates/defaultActiveOnShare loading to openai.js and responses.js controllers so shared-skill activation is respected across all agent entry points, not just initialize.js (Codex P1) - Stop keydown propagation on SkillToggle so Enter/Space does not bubble to the parent row's navigation handler (Codex P2) * fix: paginate catalog fetch and serialize toggle writes - Paginate listSkillsByAccess (up to 10 pages of 100) until the active catalog quota is filled, so inactive shared skills in recent positions do not starve active owned skills past the first page (Codex P1) - Extend listSkillsByAccess interface with cursor/has_more/after for catalog pagination - Serialize skill-state writes via a ref queue: one in-flight request at a time, with the latest desired state sent when the previous one settles. Prevents last-response-wins races where an older request overwrites newer toggles (Codex P2) * fix: share write queue across hook instances, block toggle on fetch error - Move the write queue from a per-instance useRef to a module-scoped object so every mount of useSkillActiveState (SkillList, SkillDetail, etc.) serializes against the same in-flight slot. Prior per-instance queues allowed two components to race full-map POSTs (Codex P1) - Extend the toggle guard beyond isLoading: also block when isError is true or data is undefined. Prevents a failed GET from seeding a toggle with an empty baseline that would wipe server-side overrides on the next successful POST (Codex P1) * fix: stale closure, orphan cleanup, and cap-error UX - Read toggle baseline from React Query cache via queryClient.getQueryData instead of the captured skillStates closure. The closure can be stale between onMutate's setQueryData and the next render, so rapid successive toggles would build on old state and drop earlier changes (Codex P1) - Surface the MAX_SKILL_STATES_EXCEEDED error code with a specific toast key (com_ui_skill_states_limit) so users understand the 200-cap rather than seeing a generic error - Prune orphaned entries (skillIds whose Skill doc no longer exists) on both GET and POST in SkillStatesController. Self-heals over time without needing cascade-delete hooks or a migration job. Uses one indexed Skill._id query per request * test: pin skill active-state precedence with unit tests Extract the active-state resolution logic from a closure inside injectSkillCatalog into an exported resolveSkillActive helper, then cover every branch of the precedence matrix: - Fails closed when userId is absent (even with defaultActiveOnShare=true) - Explicit override wins over ownership and config (both true and false) - Owned skills default to active when no override is set - Shared skills default to defaultActiveOnShare value - Undefined skillStates behaves identically to an empty object - defaultActiveOnShare defaults to false when omitted - Owned skills ignore defaultActiveOnShare entirely Closes Finding #2 from the pre-rebase comprehensive review. Mirrors the existing scopeSkillIds test style; injectSkillCatalog now calls resolveSkillActive instead of inlining the closure. * refactor: limit skill active toggle to detail header, drop label - Remove the per-row toggle from SkillListItem and the active-state plumbing (hook call, isSkillEnabled/onToggleEnabled/toggleAriaLabel props) from SkillList. The detail view is now the single place to change a skill's active state - Drop dim/muted styling for inactive skills in the sidebar: without a control there, the visual indication has nowhere to land - Resize SkillToggle to match neighbor buttons: outer h-9 container, h-6 w-11 track with size-5 knob, no label span. The 'Active' / 'Inactive' text that accompanied the detail-view toggle is removed - Remove the now-unused label prop and tabIndex prop (the tabIndex existed only for the list-row context) from SkillToggle. Drop the onKeyDown stopPropagation for the same reason - Remove now-orphaned com_ui_skill_active / com_ui_skill_inactive translation keys * style: shrink SkillToggle track to h-5 w-9 with size-4 knob Container stays at h-9 to match neighbor button heights. The toggle track itself drops from h-6 w-11 to h-5 w-9, with a size-4 knob travelling 1.125rem on activation. Visually lighter inside the row. * fix: remove redundant skillStates entries that match the resolved default When a toggle lands on the ownership/config default, delete the key from the map instead of persisting `{id: defaultValue}`. Without this, a user toggling a skill off and back on would leave `{id: true}` for an owned skill (whose default is already true), silently consuming a slot against the 200-entry cap. Repeated round-trip toggles could exhaust the quota with zero meaningful overrides (Codex P2). Preserves the exceptions-list invariant that the runtime-resolution design depends on. * fix: prune before enforcing skill-state cap; reject non-ObjectId keys Reorder the update controller so pruneOrphans runs before the 200-cap check. Without this, a user near the cap with some orphaned entries (skills deleted since their last GET) could send a payload that would pass after pruning but gets rejected by the raw-size check first. Add a sanity cap on raw payload size (2 * MAX_SKILL_STATES) so abusive inputs do not reach the DB query, and enforce the real cap on the pruned result instead. Harden pruneOrphans: the earlier early-return path could pass non-ObjectId keys through unchanged. Now only valid ObjectIds are returned, and the Skill-model-unavailable fallback filters by format. Also add isValidObjectIdString validation at the input boundary so malformed (but otherwise non-Mongo-unsafe) keys never reach persistence (Codex P2 x2). * fix: enforce active filter at execute time, prune revoked shares, scope queue per user P1: injectSkillCatalog now returns activeSkillIds (the filtered set that appears in the catalog). initializeAgent uses that set as the stored accessibleSkillIds on the initialized agent, so getSkillByName at runtime cannot resolve a deactivated skill — even if the LLM hallucinates a name or the user invokes by direct-invocation shorthand. Previously the executor authorized against the full ACL set, bypassing the active-state guarantee (Codex P1). P2: pruneOrphans now checks user access via findAccessibleResources in addition to skill existence. When a share is revoked, the user's skillStates entry for that skill had no cleanup path and silently consumed the 200-cap. Self-heals on both GET and POST. One extra ACL query per settings read/write; scoped to a single user so no N-user amplification (Codex P2). P2: the write queue moves from a single module-scoped object to a Map keyed by userId. Logout/login in the same tab can no longer flush the previous user's pending snapshot under the new session's auth. Each userId gets its own pending/inFlight slot; the in-flight request retains its original auth via the cookie already attached when sent, so the race window closes (Codex P2). * refactor: extract skillStates helpers to packages/api; add tests; polish Address the remaining valid findings from the comprehensive review: - Extract toRecord, loadSkillStates, validateSkillStatesPayload, and pruneOrphanSkillStates into packages/api/src/skills/skillStates.ts as TypeScript. The controller in /api shrinks to a ~90-line thin wrapper that builds live dependency adapters for Mongoose + the permission service (Review #2 DRY, #3 workspace boundary) - Replace the triplicated 12-line skillStates loading block in initialize.js, openai.js, and responses.js with a single call to loadSkillStates from @librechat/api. One helper, three sites - Swap console.error for the project logger in the controller (Review #7) - Remove the redundant INVALID_KEY_PATTERN regex: a valid ObjectId cannot contain . or $, so isValidObjectIdString already covers it (Review #11) - Parameterize the 200-cap error toast with {{0}} interpolation driven by the error response's `limit` field, so future changes to MAX_SKILL_STATES update the UI message automatically (Review #12) - Add 24 unit tests for the new skillStates helpers (toRecord, resolveDefaultActiveOnShare, loadSkillStates, validateSkillStates- Payload, pruneOrphanSkillStates) covering success paths, malformed input, cap boundaries, and parallel-query behavior (Review #4) - Add 10 tests for injectSkillCatalog pagination covering empty accessible set, missing listSkillsByAccess, single-page filter, owned-vs-shared defaults, explicit-override precedence, multi-page collection, MAX_CATALOG_PAGES safety cap, early termination on has_more=false, additional_instructions injection, and fail-closed without userId (Review #5) Total test count: 60 (was 26 on this surface). * fix: rename skillStates ValidationError to avoid barrel-export collision packages/api/src/types/error.ts already exports a ValidationError (MongooseError extension). Re-exporting a different shape from skills/skillStates.ts through the skills barrel caused TS2308 in CI because the root index re-exports both. Rename to SkillStatesValidationError to keep the exports disjoint. * refactor: tighten tests and absorb caller guard into loadSkillStates Address the followup review findings: - Add optional `accessibleSkillIds` param to loadSkillStates so the helper short-circuits to defaults when no skills are accessible. All three controllers drop the residual 7-line conditional wrapper in favor of a single destructured call (Review #2) - Remove the unreachable `typeof key !== 'string'` check from validateSkillStatesPayload: Object.entries always yields string keys per the JS spec (Review #3) - Replace the two `as unknown as` agent casts in the injectSkillCatalog tests with a `makeAgent()` factory typed directly as the function's parameter shape (Review #4) - Tighten the MAX_CATALOG_PAGES assertion from `toBeLessThanOrEqual(11)` to `toHaveBeenCalledTimes(10)` — the loop deterministically makes exactly 10 page fetches before hitting the cap (Review #1) - Rewrite the parallel-execution test for pruneOrphanSkillStates using deferred promises instead of microtask-order assertions. The test now inspects `toHaveBeenCalledTimes(1)` on both mocks after a single Promise.resolve() yield, pinning Promise.all usage without relying on push-order into a shared array (Review #5) - Evict stale writeQueue entries on user change via a module-scoped `lastSeenUserId` sentinel. When a different user's toggle is the first one after a logout/login, the previous user's queue entry is deleted. Keeps the Map bounded without adding hook-instance effect cleanup (Review #6) * fix(test): mock loadSkillStates in openai and responses controller specs The prior refactor replaced the inline 12-line skillStates loading block with a call to loadSkillStates from @librechat/api. Both controller spec files mock @librechat/api as a flat object, so any new named import from that package is undefined in the test env. Calling `await loadSkillStates(...)` threw before recordCollectedUsage ran, surfacing as "undefined is not iterable" on the test's array destructure of `mockRecordCollectedUsage.mock.calls[0]`. Add the missing mock to both spec files alongside the existing scopeSkillIds stub. * fix: abandon stale skillStates write queues on user switch Close the cross-session leak window where an in-flight flush loop still holds a reference to a previous user's queue: it could fire its next mutateAsync under the new session's auth cookies and persist the stale snapshot to the new user's document (Codex P1). Add an `abandoned` flag on `WriteQueue`. Three mechanisms cooperate: - `getWriteQueue` marks every non-active queue abandoned when the user differs from the last-seen identity (pre-existing eviction site, now more aggressive). - A `useEffect` on `userId` calls the same abandonment pass on every render with a new active identity, covering the window between logout/login and the new user's first toggle (when `getWriteQueue` would otherwise not fire). - The flush loop checks `!queue.abandoned` in its while condition so the second and later iterations exit without firing another `mutateAsync` after the session changes. The first iteration's in-flight request (already dispatched under the original user's cookies) still runs to completion or failure on its own — only the subsequent iterations, which are the dangerous ones, are blocked.
fuuuzzy
pushed a commit
to fuuuzzy/LibreChat
that referenced
this pull request
May 3, 2026
…efaults (LibreChat-AI#12692) * feat: per-user skill active/inactive toggle with ownership-aware defaults - Add `skillStates` map (Record<string, boolean>) to user schema for per-user active/inactive overrides on skills - Add `defaultActiveOnShare` to interface.skills config (default: false) so admins can control whether shared skills auto-activate - Add GET/POST /api/user/settings/skills/active endpoints with validation - Add React Query hooks with optimistic mutations for skill states - Add useSkillActiveState hook with ownership-aware resolution: owned skills default active, shared skills default inactive - Add toggle switch UI to SkillListItem and SkillDetail components - Filter inactive skills in injectSkillCatalog before agent injection - Add localization keys for active/inactive labels * fix: use Record instead of Map for IUser.skillStates Mongoose .lean() flattens Map to a plain object, causing type incompatibility with IUser in methods that return lean documents. * fix: address review findings for skill active states - Fail-closed when userId is absent: filter rejects all shared skills instead of passing them through unfiltered (Codex P1) - Validate Mongoose Map key characters (reject . and $) in controller to return 400 instead of a 500 from schema validation (Codex P2) - Block toggle while initial skill states query is loading to prevent overwriting server-side overrides with an empty snapshot (Codex P2) - Extract shared SkillToggle component, eliminating duplicate toggle markup in SkillListItem and SkillDetail (Finding LibreChat-AI#3) - Move skill state query/mutation hooks from Favorites.ts to Skills/queries.ts per feature-directory convention (Finding LibreChat-AI#4) - Fix hardcoded English aria-label in SkillListItem by passing the localized string from the parent SkillList (Finding LibreChat-AI#5) - Fix inline arrow in SkillList render loop: pass stable callback reference so SkillListItem memo() is not invalidated (Finding LibreChat-AI#1) - Extract toRecord() helper in controller to DRY the Map-to-Object conversion (Finding LibreChat-AI#6) - Remove Promise.resolve wrapping synchronous config read (Finding LibreChat-AI#8) - Remove unused TUpdateSkillStatesRequest type (Finding LibreChat-AI#12) * fix: forward tabIndex on SkillToggle to preserve list keyboard nav The original inline toggle had tabIndex={-1} so the row itself remained the sole tab target. The extraction into SkillToggle dropped this prop, making every list toggle a tab stop. Add an optional tabIndex prop and pass -1 from SkillListItem. * fix: plumb skillStates to all agent entry points, isolate toggle keydown - Add skillStates/defaultActiveOnShare loading to openai.js and responses.js controllers so shared-skill activation is respected across all agent entry points, not just initialize.js (Codex P1) - Stop keydown propagation on SkillToggle so Enter/Space does not bubble to the parent row's navigation handler (Codex P2) * fix: paginate catalog fetch and serialize toggle writes - Paginate listSkillsByAccess (up to 10 pages of 100) until the active catalog quota is filled, so inactive shared skills in recent positions do not starve active owned skills past the first page (Codex P1) - Extend listSkillsByAccess interface with cursor/has_more/after for catalog pagination - Serialize skill-state writes via a ref queue: one in-flight request at a time, with the latest desired state sent when the previous one settles. Prevents last-response-wins races where an older request overwrites newer toggles (Codex P2) * fix: share write queue across hook instances, block toggle on fetch error - Move the write queue from a per-instance useRef to a module-scoped object so every mount of useSkillActiveState (SkillList, SkillDetail, etc.) serializes against the same in-flight slot. Prior per-instance queues allowed two components to race full-map POSTs (Codex P1) - Extend the toggle guard beyond isLoading: also block when isError is true or data is undefined. Prevents a failed GET from seeding a toggle with an empty baseline that would wipe server-side overrides on the next successful POST (Codex P1) * fix: stale closure, orphan cleanup, and cap-error UX - Read toggle baseline from React Query cache via queryClient.getQueryData instead of the captured skillStates closure. The closure can be stale between onMutate's setQueryData and the next render, so rapid successive toggles would build on old state and drop earlier changes (Codex P1) - Surface the MAX_SKILL_STATES_EXCEEDED error code with a specific toast key (com_ui_skill_states_limit) so users understand the 200-cap rather than seeing a generic error - Prune orphaned entries (skillIds whose Skill doc no longer exists) on both GET and POST in SkillStatesController. Self-heals over time without needing cascade-delete hooks or a migration job. Uses one indexed Skill._id query per request * test: pin skill active-state precedence with unit tests Extract the active-state resolution logic from a closure inside injectSkillCatalog into an exported resolveSkillActive helper, then cover every branch of the precedence matrix: - Fails closed when userId is absent (even with defaultActiveOnShare=true) - Explicit override wins over ownership and config (both true and false) - Owned skills default to active when no override is set - Shared skills default to defaultActiveOnShare value - Undefined skillStates behaves identically to an empty object - defaultActiveOnShare defaults to false when omitted - Owned skills ignore defaultActiveOnShare entirely Closes Finding LibreChat-AI#2 from the pre-rebase comprehensive review. Mirrors the existing scopeSkillIds test style; injectSkillCatalog now calls resolveSkillActive instead of inlining the closure. * refactor: limit skill active toggle to detail header, drop label - Remove the per-row toggle from SkillListItem and the active-state plumbing (hook call, isSkillEnabled/onToggleEnabled/toggleAriaLabel props) from SkillList. The detail view is now the single place to change a skill's active state - Drop dim/muted styling for inactive skills in the sidebar: without a control there, the visual indication has nowhere to land - Resize SkillToggle to match neighbor buttons: outer h-9 container, h-6 w-11 track with size-5 knob, no label span. The 'Active' / 'Inactive' text that accompanied the detail-view toggle is removed - Remove the now-unused label prop and tabIndex prop (the tabIndex existed only for the list-row context) from SkillToggle. Drop the onKeyDown stopPropagation for the same reason - Remove now-orphaned com_ui_skill_active / com_ui_skill_inactive translation keys * style: shrink SkillToggle track to h-5 w-9 with size-4 knob Container stays at h-9 to match neighbor button heights. The toggle track itself drops from h-6 w-11 to h-5 w-9, with a size-4 knob travelling 1.125rem on activation. Visually lighter inside the row. * fix: remove redundant skillStates entries that match the resolved default When a toggle lands on the ownership/config default, delete the key from the map instead of persisting `{id: defaultValue}`. Without this, a user toggling a skill off and back on would leave `{id: true}` for an owned skill (whose default is already true), silently consuming a slot against the 200-entry cap. Repeated round-trip toggles could exhaust the quota with zero meaningful overrides (Codex P2). Preserves the exceptions-list invariant that the runtime-resolution design depends on. * fix: prune before enforcing skill-state cap; reject non-ObjectId keys Reorder the update controller so pruneOrphans runs before the 200-cap check. Without this, a user near the cap with some orphaned entries (skills deleted since their last GET) could send a payload that would pass after pruning but gets rejected by the raw-size check first. Add a sanity cap on raw payload size (2 * MAX_SKILL_STATES) so abusive inputs do not reach the DB query, and enforce the real cap on the pruned result instead. Harden pruneOrphans: the earlier early-return path could pass non-ObjectId keys through unchanged. Now only valid ObjectIds are returned, and the Skill-model-unavailable fallback filters by format. Also add isValidObjectIdString validation at the input boundary so malformed (but otherwise non-Mongo-unsafe) keys never reach persistence (Codex P2 x2). * fix: enforce active filter at execute time, prune revoked shares, scope queue per user P1: injectSkillCatalog now returns activeSkillIds (the filtered set that appears in the catalog). initializeAgent uses that set as the stored accessibleSkillIds on the initialized agent, so getSkillByName at runtime cannot resolve a deactivated skill — even if the LLM hallucinates a name or the user invokes by direct-invocation shorthand. Previously the executor authorized against the full ACL set, bypassing the active-state guarantee (Codex P1). P2: pruneOrphans now checks user access via findAccessibleResources in addition to skill existence. When a share is revoked, the user's skillStates entry for that skill had no cleanup path and silently consumed the 200-cap. Self-heals on both GET and POST. One extra ACL query per settings read/write; scoped to a single user so no N-user amplification (Codex P2). P2: the write queue moves from a single module-scoped object to a Map keyed by userId. Logout/login in the same tab can no longer flush the previous user's pending snapshot under the new session's auth. Each userId gets its own pending/inFlight slot; the in-flight request retains its original auth via the cookie already attached when sent, so the race window closes (Codex P2). * refactor: extract skillStates helpers to packages/api; add tests; polish Address the remaining valid findings from the comprehensive review: - Extract toRecord, loadSkillStates, validateSkillStatesPayload, and pruneOrphanSkillStates into packages/api/src/skills/skillStates.ts as TypeScript. The controller in /api shrinks to a ~90-line thin wrapper that builds live dependency adapters for Mongoose + the permission service (Review LibreChat-AI#2 DRY, LibreChat-AI#3 workspace boundary) - Replace the triplicated 12-line skillStates loading block in initialize.js, openai.js, and responses.js with a single call to loadSkillStates from @librechat/api. One helper, three sites - Swap console.error for the project logger in the controller (Review LibreChat-AI#7) - Remove the redundant INVALID_KEY_PATTERN regex: a valid ObjectId cannot contain . or $, so isValidObjectIdString already covers it (Review LibreChat-AI#11) - Parameterize the 200-cap error toast with {{0}} interpolation driven by the error response's `limit` field, so future changes to MAX_SKILL_STATES update the UI message automatically (Review LibreChat-AI#12) - Add 24 unit tests for the new skillStates helpers (toRecord, resolveDefaultActiveOnShare, loadSkillStates, validateSkillStates- Payload, pruneOrphanSkillStates) covering success paths, malformed input, cap boundaries, and parallel-query behavior (Review LibreChat-AI#4) - Add 10 tests for injectSkillCatalog pagination covering empty accessible set, missing listSkillsByAccess, single-page filter, owned-vs-shared defaults, explicit-override precedence, multi-page collection, MAX_CATALOG_PAGES safety cap, early termination on has_more=false, additional_instructions injection, and fail-closed without userId (Review LibreChat-AI#5) Total test count: 60 (was 26 on this surface). * fix: rename skillStates ValidationError to avoid barrel-export collision packages/api/src/types/error.ts already exports a ValidationError (MongooseError extension). Re-exporting a different shape from skills/skillStates.ts through the skills barrel caused TS2308 in CI because the root index re-exports both. Rename to SkillStatesValidationError to keep the exports disjoint. * refactor: tighten tests and absorb caller guard into loadSkillStates Address the followup review findings: - Add optional `accessibleSkillIds` param to loadSkillStates so the helper short-circuits to defaults when no skills are accessible. All three controllers drop the residual 7-line conditional wrapper in favor of a single destructured call (Review LibreChat-AI#2) - Remove the unreachable `typeof key !== 'string'` check from validateSkillStatesPayload: Object.entries always yields string keys per the JS spec (Review LibreChat-AI#3) - Replace the two `as unknown as` agent casts in the injectSkillCatalog tests with a `makeAgent()` factory typed directly as the function's parameter shape (Review LibreChat-AI#4) - Tighten the MAX_CATALOG_PAGES assertion from `toBeLessThanOrEqual(11)` to `toHaveBeenCalledTimes(10)` — the loop deterministically makes exactly 10 page fetches before hitting the cap (Review LibreChat-AI#1) - Rewrite the parallel-execution test for pruneOrphanSkillStates using deferred promises instead of microtask-order assertions. The test now inspects `toHaveBeenCalledTimes(1)` on both mocks after a single Promise.resolve() yield, pinning Promise.all usage without relying on push-order into a shared array (Review LibreChat-AI#5) - Evict stale writeQueue entries on user change via a module-scoped `lastSeenUserId` sentinel. When a different user's toggle is the first one after a logout/login, the previous user's queue entry is deleted. Keeps the Map bounded without adding hook-instance effect cleanup (Review LibreChat-AI#6) * fix(test): mock loadSkillStates in openai and responses controller specs The prior refactor replaced the inline 12-line skillStates loading block with a call to loadSkillStates from @librechat/api. Both controller spec files mock @librechat/api as a flat object, so any new named import from that package is undefined in the test env. Calling `await loadSkillStates(...)` threw before recordCollectedUsage ran, surfacing as "undefined is not iterable" on the test's array destructure of `mockRecordCollectedUsage.mock.calls[0]`. Add the missing mock to both spec files alongside the existing scopeSkillIds stub. * fix: abandon stale skillStates write queues on user switch Close the cross-session leak window where an in-flight flush loop still holds a reference to a previous user's queue: it could fire its next mutateAsync under the new session's auth cookies and persist the stale snapshot to the new user's document (Codex P1). Add an `abandoned` flag on `WriteQueue`. Three mechanisms cooperate: - `getWriteQueue` marks every non-active queue abandoned when the user differs from the last-seen identity (pre-existing eviction site, now more aggressive). - A `useEffect` on `userId` calls the same abandonment pass on every render with a new active identity, covering the window between logout/login and the new user's first toggle (when `getWriteQueue` would otherwise not fire). - The flush loop checks `!queue.abandoned` in its while condition so the second and later iterations exit without firing another `mutateAsync` after the session changes. The first iteration's in-flight request (already dispatched under the original user's cookies) still runs to completion or failure on its own — only the subsequent iterations, which are the dangerous ones, are blocked.
jcbartle
pushed a commit
to jcbartle/LibreChat
that referenced
this pull request
May 11, 2026
…efaults (LibreChat-AI#12692) * feat: per-user skill active/inactive toggle with ownership-aware defaults - Add `skillStates` map (Record<string, boolean>) to user schema for per-user active/inactive overrides on skills - Add `defaultActiveOnShare` to interface.skills config (default: false) so admins can control whether shared skills auto-activate - Add GET/POST /api/user/settings/skills/active endpoints with validation - Add React Query hooks with optimistic mutations for skill states - Add useSkillActiveState hook with ownership-aware resolution: owned skills default active, shared skills default inactive - Add toggle switch UI to SkillListItem and SkillDetail components - Filter inactive skills in injectSkillCatalog before agent injection - Add localization keys for active/inactive labels * fix: use Record instead of Map for IUser.skillStates Mongoose .lean() flattens Map to a plain object, causing type incompatibility with IUser in methods that return lean documents. * fix: address review findings for skill active states - Fail-closed when userId is absent: filter rejects all shared skills instead of passing them through unfiltered (Codex P1) - Validate Mongoose Map key characters (reject . and $) in controller to return 400 instead of a 500 from schema validation (Codex P2) - Block toggle while initial skill states query is loading to prevent overwriting server-side overrides with an empty snapshot (Codex P2) - Extract shared SkillToggle component, eliminating duplicate toggle markup in SkillListItem and SkillDetail (Finding LibreChat-AI#3) - Move skill state query/mutation hooks from Favorites.ts to Skills/queries.ts per feature-directory convention (Finding LibreChat-AI#4) - Fix hardcoded English aria-label in SkillListItem by passing the localized string from the parent SkillList (Finding LibreChat-AI#5) - Fix inline arrow in SkillList render loop: pass stable callback reference so SkillListItem memo() is not invalidated (Finding LibreChat-AI#1) - Extract toRecord() helper in controller to DRY the Map-to-Object conversion (Finding LibreChat-AI#6) - Remove Promise.resolve wrapping synchronous config read (Finding LibreChat-AI#8) - Remove unused TUpdateSkillStatesRequest type (Finding LibreChat-AI#12) * fix: forward tabIndex on SkillToggle to preserve list keyboard nav The original inline toggle had tabIndex={-1} so the row itself remained the sole tab target. The extraction into SkillToggle dropped this prop, making every list toggle a tab stop. Add an optional tabIndex prop and pass -1 from SkillListItem. * fix: plumb skillStates to all agent entry points, isolate toggle keydown - Add skillStates/defaultActiveOnShare loading to openai.js and responses.js controllers so shared-skill activation is respected across all agent entry points, not just initialize.js (Codex P1) - Stop keydown propagation on SkillToggle so Enter/Space does not bubble to the parent row's navigation handler (Codex P2) * fix: paginate catalog fetch and serialize toggle writes - Paginate listSkillsByAccess (up to 10 pages of 100) until the active catalog quota is filled, so inactive shared skills in recent positions do not starve active owned skills past the first page (Codex P1) - Extend listSkillsByAccess interface with cursor/has_more/after for catalog pagination - Serialize skill-state writes via a ref queue: one in-flight request at a time, with the latest desired state sent when the previous one settles. Prevents last-response-wins races where an older request overwrites newer toggles (Codex P2) * fix: share write queue across hook instances, block toggle on fetch error - Move the write queue from a per-instance useRef to a module-scoped object so every mount of useSkillActiveState (SkillList, SkillDetail, etc.) serializes against the same in-flight slot. Prior per-instance queues allowed two components to race full-map POSTs (Codex P1) - Extend the toggle guard beyond isLoading: also block when isError is true or data is undefined. Prevents a failed GET from seeding a toggle with an empty baseline that would wipe server-side overrides on the next successful POST (Codex P1) * fix: stale closure, orphan cleanup, and cap-error UX - Read toggle baseline from React Query cache via queryClient.getQueryData instead of the captured skillStates closure. The closure can be stale between onMutate's setQueryData and the next render, so rapid successive toggles would build on old state and drop earlier changes (Codex P1) - Surface the MAX_SKILL_STATES_EXCEEDED error code with a specific toast key (com_ui_skill_states_limit) so users understand the 200-cap rather than seeing a generic error - Prune orphaned entries (skillIds whose Skill doc no longer exists) on both GET and POST in SkillStatesController. Self-heals over time without needing cascade-delete hooks or a migration job. Uses one indexed Skill._id query per request * test: pin skill active-state precedence with unit tests Extract the active-state resolution logic from a closure inside injectSkillCatalog into an exported resolveSkillActive helper, then cover every branch of the precedence matrix: - Fails closed when userId is absent (even with defaultActiveOnShare=true) - Explicit override wins over ownership and config (both true and false) - Owned skills default to active when no override is set - Shared skills default to defaultActiveOnShare value - Undefined skillStates behaves identically to an empty object - defaultActiveOnShare defaults to false when omitted - Owned skills ignore defaultActiveOnShare entirely Closes Finding LibreChat-AI#2 from the pre-rebase comprehensive review. Mirrors the existing scopeSkillIds test style; injectSkillCatalog now calls resolveSkillActive instead of inlining the closure. * refactor: limit skill active toggle to detail header, drop label - Remove the per-row toggle from SkillListItem and the active-state plumbing (hook call, isSkillEnabled/onToggleEnabled/toggleAriaLabel props) from SkillList. The detail view is now the single place to change a skill's active state - Drop dim/muted styling for inactive skills in the sidebar: without a control there, the visual indication has nowhere to land - Resize SkillToggle to match neighbor buttons: outer h-9 container, h-6 w-11 track with size-5 knob, no label span. The 'Active' / 'Inactive' text that accompanied the detail-view toggle is removed - Remove the now-unused label prop and tabIndex prop (the tabIndex existed only for the list-row context) from SkillToggle. Drop the onKeyDown stopPropagation for the same reason - Remove now-orphaned com_ui_skill_active / com_ui_skill_inactive translation keys * style: shrink SkillToggle track to h-5 w-9 with size-4 knob Container stays at h-9 to match neighbor button heights. The toggle track itself drops from h-6 w-11 to h-5 w-9, with a size-4 knob travelling 1.125rem on activation. Visually lighter inside the row. * fix: remove redundant skillStates entries that match the resolved default When a toggle lands on the ownership/config default, delete the key from the map instead of persisting `{id: defaultValue}`. Without this, a user toggling a skill off and back on would leave `{id: true}` for an owned skill (whose default is already true), silently consuming a slot against the 200-entry cap. Repeated round-trip toggles could exhaust the quota with zero meaningful overrides (Codex P2). Preserves the exceptions-list invariant that the runtime-resolution design depends on. * fix: prune before enforcing skill-state cap; reject non-ObjectId keys Reorder the update controller so pruneOrphans runs before the 200-cap check. Without this, a user near the cap with some orphaned entries (skills deleted since their last GET) could send a payload that would pass after pruning but gets rejected by the raw-size check first. Add a sanity cap on raw payload size (2 * MAX_SKILL_STATES) so abusive inputs do not reach the DB query, and enforce the real cap on the pruned result instead. Harden pruneOrphans: the earlier early-return path could pass non-ObjectId keys through unchanged. Now only valid ObjectIds are returned, and the Skill-model-unavailable fallback filters by format. Also add isValidObjectIdString validation at the input boundary so malformed (but otherwise non-Mongo-unsafe) keys never reach persistence (Codex P2 x2). * fix: enforce active filter at execute time, prune revoked shares, scope queue per user P1: injectSkillCatalog now returns activeSkillIds (the filtered set that appears in the catalog). initializeAgent uses that set as the stored accessibleSkillIds on the initialized agent, so getSkillByName at runtime cannot resolve a deactivated skill — even if the LLM hallucinates a name or the user invokes by direct-invocation shorthand. Previously the executor authorized against the full ACL set, bypassing the active-state guarantee (Codex P1). P2: pruneOrphans now checks user access via findAccessibleResources in addition to skill existence. When a share is revoked, the user's skillStates entry for that skill had no cleanup path and silently consumed the 200-cap. Self-heals on both GET and POST. One extra ACL query per settings read/write; scoped to a single user so no N-user amplification (Codex P2). P2: the write queue moves from a single module-scoped object to a Map keyed by userId. Logout/login in the same tab can no longer flush the previous user's pending snapshot under the new session's auth. Each userId gets its own pending/inFlight slot; the in-flight request retains its original auth via the cookie already attached when sent, so the race window closes (Codex P2). * refactor: extract skillStates helpers to packages/api; add tests; polish Address the remaining valid findings from the comprehensive review: - Extract toRecord, loadSkillStates, validateSkillStatesPayload, and pruneOrphanSkillStates into packages/api/src/skills/skillStates.ts as TypeScript. The controller in /api shrinks to a ~90-line thin wrapper that builds live dependency adapters for Mongoose + the permission service (Review LibreChat-AI#2 DRY, LibreChat-AI#3 workspace boundary) - Replace the triplicated 12-line skillStates loading block in initialize.js, openai.js, and responses.js with a single call to loadSkillStates from @librechat/api. One helper, three sites - Swap console.error for the project logger in the controller (Review LibreChat-AI#7) - Remove the redundant INVALID_KEY_PATTERN regex: a valid ObjectId cannot contain . or $, so isValidObjectIdString already covers it (Review LibreChat-AI#11) - Parameterize the 200-cap error toast with {{0}} interpolation driven by the error response's `limit` field, so future changes to MAX_SKILL_STATES update the UI message automatically (Review LibreChat-AI#12) - Add 24 unit tests for the new skillStates helpers (toRecord, resolveDefaultActiveOnShare, loadSkillStates, validateSkillStates- Payload, pruneOrphanSkillStates) covering success paths, malformed input, cap boundaries, and parallel-query behavior (Review LibreChat-AI#4) - Add 10 tests for injectSkillCatalog pagination covering empty accessible set, missing listSkillsByAccess, single-page filter, owned-vs-shared defaults, explicit-override precedence, multi-page collection, MAX_CATALOG_PAGES safety cap, early termination on has_more=false, additional_instructions injection, and fail-closed without userId (Review LibreChat-AI#5) Total test count: 60 (was 26 on this surface). * fix: rename skillStates ValidationError to avoid barrel-export collision packages/api/src/types/error.ts already exports a ValidationError (MongooseError extension). Re-exporting a different shape from skills/skillStates.ts through the skills barrel caused TS2308 in CI because the root index re-exports both. Rename to SkillStatesValidationError to keep the exports disjoint. * refactor: tighten tests and absorb caller guard into loadSkillStates Address the followup review findings: - Add optional `accessibleSkillIds` param to loadSkillStates so the helper short-circuits to defaults when no skills are accessible. All three controllers drop the residual 7-line conditional wrapper in favor of a single destructured call (Review LibreChat-AI#2) - Remove the unreachable `typeof key !== 'string'` check from validateSkillStatesPayload: Object.entries always yields string keys per the JS spec (Review LibreChat-AI#3) - Replace the two `as unknown as` agent casts in the injectSkillCatalog tests with a `makeAgent()` factory typed directly as the function's parameter shape (Review LibreChat-AI#4) - Tighten the MAX_CATALOG_PAGES assertion from `toBeLessThanOrEqual(11)` to `toHaveBeenCalledTimes(10)` — the loop deterministically makes exactly 10 page fetches before hitting the cap (Review LibreChat-AI#1) - Rewrite the parallel-execution test for pruneOrphanSkillStates using deferred promises instead of microtask-order assertions. The test now inspects `toHaveBeenCalledTimes(1)` on both mocks after a single Promise.resolve() yield, pinning Promise.all usage without relying on push-order into a shared array (Review LibreChat-AI#5) - Evict stale writeQueue entries on user change via a module-scoped `lastSeenUserId` sentinel. When a different user's toggle is the first one after a logout/login, the previous user's queue entry is deleted. Keeps the Map bounded without adding hook-instance effect cleanup (Review LibreChat-AI#6) * fix(test): mock loadSkillStates in openai and responses controller specs The prior refactor replaced the inline 12-line skillStates loading block with a call to loadSkillStates from @librechat/api. Both controller spec files mock @librechat/api as a flat object, so any new named import from that package is undefined in the test env. Calling `await loadSkillStates(...)` threw before recordCollectedUsage ran, surfacing as "undefined is not iterable" on the test's array destructure of `mockRecordCollectedUsage.mock.calls[0]`. Add the missing mock to both spec files alongside the existing scopeSkillIds stub. * fix: abandon stale skillStates write queues on user switch Close the cross-session leak window where an in-flight flush loop still holds a reference to a previous user's queue: it could fire its next mutateAsync under the new session's auth cookies and persist the stale snapshot to the new user's document (Codex P1). Add an `abandoned` flag on `WriteQueue`. Three mechanisms cooperate: - `getWriteQueue` marks every non-active queue abandoned when the user differs from the last-seen identity (pre-existing eviction site, now more aggressive). - A `useEffect` on `userId` calls the same abandonment pass on every render with a new active identity, covering the window between logout/login and the new user's first toggle (when `getWriteQueue` would otherwise not fire). - The flush loop checks `!queue.abandoned` in its while condition so the second and later iterations exit without firing another `mutateAsync` after the session changes. The first iteration's in-flight request (already dispatched under the original user's cookies) still runs to completion or failure on its own — only the subsequent iterations, which are the dangerous ones, are blocked.
ThomasVuNguyen
pushed a commit
to ThomasVuNguyen/LibreChat
that referenced
this pull request
Jul 15, 2026
…efaults (LibreChat-AI#12692) * feat: per-user skill active/inactive toggle with ownership-aware defaults - Add `skillStates` map (Record<string, boolean>) to user schema for per-user active/inactive overrides on skills - Add `defaultActiveOnShare` to interface.skills config (default: false) so admins can control whether shared skills auto-activate - Add GET/POST /api/user/settings/skills/active endpoints with validation - Add React Query hooks with optimistic mutations for skill states - Add useSkillActiveState hook with ownership-aware resolution: owned skills default active, shared skills default inactive - Add toggle switch UI to SkillListItem and SkillDetail components - Filter inactive skills in injectSkillCatalog before agent injection - Add localization keys for active/inactive labels * fix: use Record instead of Map for IUser.skillStates Mongoose .lean() flattens Map to a plain object, causing type incompatibility with IUser in methods that return lean documents. * fix: address review findings for skill active states - Fail-closed when userId is absent: filter rejects all shared skills instead of passing them through unfiltered (Codex P1) - Validate Mongoose Map key characters (reject . and $) in controller to return 400 instead of a 500 from schema validation (Codex P2) - Block toggle while initial skill states query is loading to prevent overwriting server-side overrides with an empty snapshot (Codex P2) - Extract shared SkillToggle component, eliminating duplicate toggle markup in SkillListItem and SkillDetail (Finding LibreChat-AI#3) - Move skill state query/mutation hooks from Favorites.ts to Skills/queries.ts per feature-directory convention (Finding LibreChat-AI#4) - Fix hardcoded English aria-label in SkillListItem by passing the localized string from the parent SkillList (Finding LibreChat-AI#5) - Fix inline arrow in SkillList render loop: pass stable callback reference so SkillListItem memo() is not invalidated (Finding #1) - Extract toRecord() helper in controller to DRY the Map-to-Object conversion (Finding LibreChat-AI#6) - Remove Promise.resolve wrapping synchronous config read (Finding LibreChat-AI#8) - Remove unused TUpdateSkillStatesRequest type (Finding LibreChat-AI#12) * fix: forward tabIndex on SkillToggle to preserve list keyboard nav The original inline toggle had tabIndex={-1} so the row itself remained the sole tab target. The extraction into SkillToggle dropped this prop, making every list toggle a tab stop. Add an optional tabIndex prop and pass -1 from SkillListItem. * fix: plumb skillStates to all agent entry points, isolate toggle keydown - Add skillStates/defaultActiveOnShare loading to openai.js and responses.js controllers so shared-skill activation is respected across all agent entry points, not just initialize.js (Codex P1) - Stop keydown propagation on SkillToggle so Enter/Space does not bubble to the parent row's navigation handler (Codex P2) * fix: paginate catalog fetch and serialize toggle writes - Paginate listSkillsByAccess (up to 10 pages of 100) until the active catalog quota is filled, so inactive shared skills in recent positions do not starve active owned skills past the first page (Codex P1) - Extend listSkillsByAccess interface with cursor/has_more/after for catalog pagination - Serialize skill-state writes via a ref queue: one in-flight request at a time, with the latest desired state sent when the previous one settles. Prevents last-response-wins races where an older request overwrites newer toggles (Codex P2) * fix: share write queue across hook instances, block toggle on fetch error - Move the write queue from a per-instance useRef to a module-scoped object so every mount of useSkillActiveState (SkillList, SkillDetail, etc.) serializes against the same in-flight slot. Prior per-instance queues allowed two components to race full-map POSTs (Codex P1) - Extend the toggle guard beyond isLoading: also block when isError is true or data is undefined. Prevents a failed GET from seeding a toggle with an empty baseline that would wipe server-side overrides on the next successful POST (Codex P1) * fix: stale closure, orphan cleanup, and cap-error UX - Read toggle baseline from React Query cache via queryClient.getQueryData instead of the captured skillStates closure. The closure can be stale between onMutate's setQueryData and the next render, so rapid successive toggles would build on old state and drop earlier changes (Codex P1) - Surface the MAX_SKILL_STATES_EXCEEDED error code with a specific toast key (com_ui_skill_states_limit) so users understand the 200-cap rather than seeing a generic error - Prune orphaned entries (skillIds whose Skill doc no longer exists) on both GET and POST in SkillStatesController. Self-heals over time without needing cascade-delete hooks or a migration job. Uses one indexed Skill._id query per request * test: pin skill active-state precedence with unit tests Extract the active-state resolution logic from a closure inside injectSkillCatalog into an exported resolveSkillActive helper, then cover every branch of the precedence matrix: - Fails closed when userId is absent (even with defaultActiveOnShare=true) - Explicit override wins over ownership and config (both true and false) - Owned skills default to active when no override is set - Shared skills default to defaultActiveOnShare value - Undefined skillStates behaves identically to an empty object - defaultActiveOnShare defaults to false when omitted - Owned skills ignore defaultActiveOnShare entirely Closes Finding #2 from the pre-rebase comprehensive review. Mirrors the existing scopeSkillIds test style; injectSkillCatalog now calls resolveSkillActive instead of inlining the closure. * refactor: limit skill active toggle to detail header, drop label - Remove the per-row toggle from SkillListItem and the active-state plumbing (hook call, isSkillEnabled/onToggleEnabled/toggleAriaLabel props) from SkillList. The detail view is now the single place to change a skill's active state - Drop dim/muted styling for inactive skills in the sidebar: without a control there, the visual indication has nowhere to land - Resize SkillToggle to match neighbor buttons: outer h-9 container, h-6 w-11 track with size-5 knob, no label span. The 'Active' / 'Inactive' text that accompanied the detail-view toggle is removed - Remove the now-unused label prop and tabIndex prop (the tabIndex existed only for the list-row context) from SkillToggle. Drop the onKeyDown stopPropagation for the same reason - Remove now-orphaned com_ui_skill_active / com_ui_skill_inactive translation keys * style: shrink SkillToggle track to h-5 w-9 with size-4 knob Container stays at h-9 to match neighbor button heights. The toggle track itself drops from h-6 w-11 to h-5 w-9, with a size-4 knob travelling 1.125rem on activation. Visually lighter inside the row. * fix: remove redundant skillStates entries that match the resolved default When a toggle lands on the ownership/config default, delete the key from the map instead of persisting `{id: defaultValue}`. Without this, a user toggling a skill off and back on would leave `{id: true}` for an owned skill (whose default is already true), silently consuming a slot against the 200-entry cap. Repeated round-trip toggles could exhaust the quota with zero meaningful overrides (Codex P2). Preserves the exceptions-list invariant that the runtime-resolution design depends on. * fix: prune before enforcing skill-state cap; reject non-ObjectId keys Reorder the update controller so pruneOrphans runs before the 200-cap check. Without this, a user near the cap with some orphaned entries (skills deleted since their last GET) could send a payload that would pass after pruning but gets rejected by the raw-size check first. Add a sanity cap on raw payload size (2 * MAX_SKILL_STATES) so abusive inputs do not reach the DB query, and enforce the real cap on the pruned result instead. Harden pruneOrphans: the earlier early-return path could pass non-ObjectId keys through unchanged. Now only valid ObjectIds are returned, and the Skill-model-unavailable fallback filters by format. Also add isValidObjectIdString validation at the input boundary so malformed (but otherwise non-Mongo-unsafe) keys never reach persistence (Codex P2 x2). * fix: enforce active filter at execute time, prune revoked shares, scope queue per user P1: injectSkillCatalog now returns activeSkillIds (the filtered set that appears in the catalog). initializeAgent uses that set as the stored accessibleSkillIds on the initialized agent, so getSkillByName at runtime cannot resolve a deactivated skill — even if the LLM hallucinates a name or the user invokes by direct-invocation shorthand. Previously the executor authorized against the full ACL set, bypassing the active-state guarantee (Codex P1). P2: pruneOrphans now checks user access via findAccessibleResources in addition to skill existence. When a share is revoked, the user's skillStates entry for that skill had no cleanup path and silently consumed the 200-cap. Self-heals on both GET and POST. One extra ACL query per settings read/write; scoped to a single user so no N-user amplification (Codex P2). P2: the write queue moves from a single module-scoped object to a Map keyed by userId. Logout/login in the same tab can no longer flush the previous user's pending snapshot under the new session's auth. Each userId gets its own pending/inFlight slot; the in-flight request retains its original auth via the cookie already attached when sent, so the race window closes (Codex P2). * refactor: extract skillStates helpers to packages/api; add tests; polish Address the remaining valid findings from the comprehensive review: - Extract toRecord, loadSkillStates, validateSkillStatesPayload, and pruneOrphanSkillStates into packages/api/src/skills/skillStates.ts as TypeScript. The controller in /api shrinks to a ~90-line thin wrapper that builds live dependency adapters for Mongoose + the permission service (Review #2 DRY, LibreChat-AI#3 workspace boundary) - Replace the triplicated 12-line skillStates loading block in initialize.js, openai.js, and responses.js with a single call to loadSkillStates from @librechat/api. One helper, three sites - Swap console.error for the project logger in the controller (Review LibreChat-AI#7) - Remove the redundant INVALID_KEY_PATTERN regex: a valid ObjectId cannot contain . or $, so isValidObjectIdString already covers it (Review LibreChat-AI#11) - Parameterize the 200-cap error toast with {{0}} interpolation driven by the error response's `limit` field, so future changes to MAX_SKILL_STATES update the UI message automatically (Review LibreChat-AI#12) - Add 24 unit tests for the new skillStates helpers (toRecord, resolveDefaultActiveOnShare, loadSkillStates, validateSkillStates- Payload, pruneOrphanSkillStates) covering success paths, malformed input, cap boundaries, and parallel-query behavior (Review LibreChat-AI#4) - Add 10 tests for injectSkillCatalog pagination covering empty accessible set, missing listSkillsByAccess, single-page filter, owned-vs-shared defaults, explicit-override precedence, multi-page collection, MAX_CATALOG_PAGES safety cap, early termination on has_more=false, additional_instructions injection, and fail-closed without userId (Review LibreChat-AI#5) Total test count: 60 (was 26 on this surface). * fix: rename skillStates ValidationError to avoid barrel-export collision packages/api/src/types/error.ts already exports a ValidationError (MongooseError extension). Re-exporting a different shape from skills/skillStates.ts through the skills barrel caused TS2308 in CI because the root index re-exports both. Rename to SkillStatesValidationError to keep the exports disjoint. * refactor: tighten tests and absorb caller guard into loadSkillStates Address the followup review findings: - Add optional `accessibleSkillIds` param to loadSkillStates so the helper short-circuits to defaults when no skills are accessible. All three controllers drop the residual 7-line conditional wrapper in favor of a single destructured call (Review #2) - Remove the unreachable `typeof key !== 'string'` check from validateSkillStatesPayload: Object.entries always yields string keys per the JS spec (Review LibreChat-AI#3) - Replace the two `as unknown as` agent casts in the injectSkillCatalog tests with a `makeAgent()` factory typed directly as the function's parameter shape (Review LibreChat-AI#4) - Tighten the MAX_CATALOG_PAGES assertion from `toBeLessThanOrEqual(11)` to `toHaveBeenCalledTimes(10)` — the loop deterministically makes exactly 10 page fetches before hitting the cap (Review #1) - Rewrite the parallel-execution test for pruneOrphanSkillStates using deferred promises instead of microtask-order assertions. The test now inspects `toHaveBeenCalledTimes(1)` on both mocks after a single Promise.resolve() yield, pinning Promise.all usage without relying on push-order into a shared array (Review LibreChat-AI#5) - Evict stale writeQueue entries on user change via a module-scoped `lastSeenUserId` sentinel. When a different user's toggle is the first one after a logout/login, the previous user's queue entry is deleted. Keeps the Map bounded without adding hook-instance effect cleanup (Review LibreChat-AI#6) * fix(test): mock loadSkillStates in openai and responses controller specs The prior refactor replaced the inline 12-line skillStates loading block with a call to loadSkillStates from @librechat/api. Both controller spec files mock @librechat/api as a flat object, so any new named import from that package is undefined in the test env. Calling `await loadSkillStates(...)` threw before recordCollectedUsage ran, surfacing as "undefined is not iterable" on the test's array destructure of `mockRecordCollectedUsage.mock.calls[0]`. Add the missing mock to both spec files alongside the existing scopeSkillIds stub. * fix: abandon stale skillStates write queues on user switch Close the cross-session leak window where an in-flight flush loop still holds a reference to a previous user's queue: it could fire its next mutateAsync under the new session's auth cookies and persist the stale snapshot to the new user's document (Codex P1). Add an `abandoned` flag on `WriteQueue`. Three mechanisms cooperate: - `getWriteQueue` marks every non-active queue abandoned when the user differs from the last-seen identity (pre-existing eviction site, now more aggressive). - A `useEffect` on `userId` calls the same abandonment pass on every render with a new active identity, covering the window between logout/login and the new user's first toggle (when `getWriteQueue` would otherwise not fire). - The flush loop checks `!queue.abandoned` in its while condition so the second and later iterations exit without firing another `mutateAsync` after the session changes. The first iteration's in-flight request (already dispatched under the original user's cookies) still runs to completion or failure on its own — only the subsequent iterations, which are the dangerous ones, are blocked.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
I added a per-user skill active/inactive toggle so that shared skills require explicit opt-in before being injected into agent sessions. Owned skills remain active by default, while shared skills default to inactive, preventing surprises when someone shares a skill. An admin config override (
defaultActiveOnShare) lets administrators force shared skills to auto-activate in controlled environments.skillStatesmap (Record<string, boolean>) on the user document, tracking only explicit toggles rather than duplicating the full skill listdefaultActiveOnShareboolean to theinterface.skillsZod config schema (defaults tofalse), surfaced inlibrechat.yamland startup config for admin controlGET/POST /api/user/settings/skills/activeendpoints inSkillStatesController.jswith validation (max 200 entries, key length, boolean values)useGetSkillStatesQueryanduseUpdateSkillStatesMutationReact Query hooks with optimistic updates and rollback on error, following the existing favorites patternuseSkillActiveStatehook with ownership-aware resolution: checks explicit override first, then falls back to owned=active / shared=defaultActiveOnShareSkillListItemrows (with dimmed icon/text when inactive) and to theSkillDetailheader alongside Share/Edit/Delete actionsinjectSkillCatalogafterlistSkillsByAccessreturns, before building the agent catalog, so inactive skills never reach the LLMuserId,skillStates, anddefaultActiveOnSharethrough theinitializeAgentchain for both primary and secondary (handoff) agentscom_ui_skill_active,com_ui_skill_inactive, andcom_ui_skill_toggle_activelocalization keysChange Type
Testing
interface.skills.defaultActiveOnShare: trueinlibrechat.yaml, restart, and verify shared skills now auto-activate for users who have not explicitly toggled themSkillDetailview matches the sidebar toggle stateTest Configuration:
librechat.yamlwith skills capability enabledChecklist