Repository navigation
feat: load project skills in chats - #30838
Conversation
Add sqlc queries for skills owned by a chat project and authorize them through the parent project in dbauthz: reading needs chat_project:read and writing needs chat_project:update. Skill.RBACObject keeps denying project rows so any unwrapped path fails closed. Add the chat_project_skill audit resource type and AuditableChatProjectSkill so the upcoming project skill API can record audit entries.
Add experimental routes to list, create, read, update and delete skills on a chat project. Reading needs read on the project and writes need update; callers who lack the permission get a 404 like the project instructions routes. The handlers reuse the shared skill create and update path, map the per-project cap and name index to 409, and audit every write as a chat_project_skill. Add matching ExperimentalClient methods.
Chats in a project now get the project's enabled skills as a new "project" skill source, read as the chat owner each turn. Subagents use the root chat's project. read_skill loads project skill bodies, and MergeSkills takes sources in merge order so new sources do not change its signature.
The "/" menu offers the skills of the chat's project, next to personal, organization and workspace skills, and /clear and /compact defer to a project skill with the same name. Documents project skills.
…g in the slash menu Organization and project skills load through one hook, and the menu renders its skill groups from one ordered list. Adds project loading and error stories.
…lify skill group offsets
…nd resolve emptiness in the hook
…naming in the composer
Cover the project-only skill tool check in the integration test, the project skill body loader's not-found mapping, and the chat-projects experiment gate in the composer.
Docs previewCheck off each page once it's been reviewed. If a page changes in a later push, its checkbox clears automatically so it gets a fresh look. Pages not yet wired into the docs navigation aren't listed here. |
|
@codex review |
Codex Review SummaryThis comment shows the latest Codex review activity on this pull request.
ℹ️ About Codex in GitHubYour team has set up Codex to review pull requests in this repo. Reviews are triggered when you
Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings. |
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 681b37465f
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".
… request instead of menu markup
|
@codex review |
|
Codex Review: Didn't find any major issues. Nice work! Reviewed commit: ℹ️ About Codex in GitHubYour team has set up Codex to review pull requests in this repo. Reviews are triggered when you
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". |
Every observer of the project skills query now shares its retry policy, and a 403 or 404 after an earlier load also counts as no project skills.
…st access after project skills loaded
|
@codex review |
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 29646530af
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".
chatd loads project skills only when the chat owner owns the project or it is shared with them, while the skills list endpoint also honors role grants. The page and the composer now also consult the caller's project list, which holds only owned and shared projects, and keep project skills unknown while that list is unsettled. The composer tests mock the project list, seed caches with the exported chatProjectSkillsKey, and wait on query state instead of DOM absence.
|
@codex review |
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 89a2e9a0a2
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".
…il refetched A cached list can predate a share's revocation as well as its creation, so a project it contains is no more current than one it lacks.
|
@codex review |
|
Codex Review: Didn't find any major issues. Another round soon, please! Reviewed commit: ℹ️ About Codex in GitHubYour team has set up Codex to review pull requests in this repo. Reviews are triggered when you
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". |
Turns check project access once with resolveUsableChatProjectID and then load instructions and skills, so resolveProjectInstructions had no production callers. Its tests now cover resolveUsableChatProjectID and loadProjectInstructions directly.
|
@codex review |
|
Codex Review: Didn't find any major issues. 🚀 Reviewed commit: ℹ️ About Codex in GitHubYour team has set up Codex to review pull requests in this repo. Reviews are triggered when you
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". |
…coder into chat-project-skills/chatd
…retry prompt for unusable projects A 403 or 404 from the project skills request means the caller lost access to the project, which the composer already treats as no project skills. The slash menu still showed "Could not load project skills" with a retry hint that could never succeed, so a project the caller cannot use no longer shows a loading or error status. A story covers the menu in that state.
|
@codex review |
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: e9a4adf7c6
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".
| }, | ||
| parameters: { | ||
| queries: [ | ||
| { key: organizationSkills(mockOrganizationId).queryKey, data: [] }, |
There was a problem hiding this comment.
Export and use the organization-skills query key
Export organizationSkillsKey and import it here instead of constructing the full query-options object solely to read .queryKey. The current setup couples the story to unrelated fetch configuration and bypasses the repository's canonical query-key pattern.
AGENTS.md reference: site/AGENTS.md:L13-L13
Useful? React with 👍 / 👎.
| const projectSkillsQuery = useQuery({ | ||
| ...chatProjectSkills(chatOrganizationId, chatProjectId ?? ""), | ||
| enabled: hasChatProject, | ||
| staleTime: 60_000, | ||
| }); |
There was a problem hiding this comment.
Extract the page-level project-skill orchestration
Move the added project-skills queries and usability derivation into a focused hook. Adding this request and state orchestration directly to an already roughly 870-line page makes the component harder to navigate and exceeds the repository's component-size guidance.
AGENTS.md reference: site/AGENTS.md:L46-L46
Useful? React with 👍 / 👎.
| const projectSource = useStoredSkillSource( | ||
| chatProjectSkills(organizationId ?? "", projectId ?? ""), | ||
| { hasSource: hasProject, isTriggerActive: hasSkillsTrigger }, | ||
| ); | ||
| const projectQuery = useQuery({ |
There was a problem hiding this comment.
Extract the composer project-skill state pipeline
Move the project queries and their loading, error, usability, and filtering derivations into a focused hook. This block adds substantial server-state orchestration to an already roughly 1,167-line editor component, making the file increasingly difficult to maintain.
AGENTS.md reference: site/AGENTS.md:L46-L46
Useful? React with 👍 / 👎.
| }, | ||
| play: async ({ canvasElement }) => { | ||
| await typeInEditor(canvasElement, "/"); | ||
| // Built-in commands appear once the project source has resolved. |
There was a problem hiding this comment.
Remove the redundant wait narration
Delete this comment because it only restates that the immediately following wait looks for the built-in command; the story-level comment already explains why this state matters.
AGENTS.md reference: site/AGENTS.md:L10-L10
Useful? React with 👍 / 👎.
Chats that belong to a chat project now get the project's skills, next to personal, organization and workspace skills. They show up in the model's skill list, load through
read_skill, and appear in a "Project skills" group in the/menu.coderd/x/skillsadds theprojectsource.MergeSkillsnow takes a list of sources in merge order (personal, organization, project, workspace). Bare names stay bare unless two sources share a name, in which case they getproject/<name>style aliases.usableChatProjectcheck that project instructions and memory use, and one check per turn covers both instructions and skills. Disabled skills are skipped. A failed load logs a warning and the turn continues without them. Subagents use their root chat's project; that lookup is now a helper shared with project instructions.chat-projectsexperiment./clearand/compactdefer to a project skill with the same name, as they already do for other sources. When the user can't use the project, the site treats it as no project skills, like chatd, so those commands keep working and the/menu shows no retry prompt for project skills. That covers a 403 or 404 from the project skills request and a project missing from the user's project list, which holds only owned and shared projects, so an administrator role that can read the project doesn't count. Other failures are retried and leave the list unknown. A cached project list is also unknown while it refetches or after a failed refetch, so a share revoked since it was cached keeps those commands pending instead of deferring to stale project skills.coderd/x/chatd/ARCHITECTURE.mdgains a paragraph on project skills.Dogfood UAT on this branch passed: project skills appear in the project page
/menu and in project chats,read_skillloads them, disabled and deleted skills report not found, chats outside the project don't see them, and a subagent gets them through its root chat. The site reads project skills from the chat's ownproject_id, so subagent chats show them in the composer once sub-chats carry their root's project (#30683)./menuread_skillin a project chatStack (merge top to bottom):