-
Notifications
You must be signed in to change notification settings - Fork 0
Expand file tree
/
Copy pathupdates.yaml
More file actions
71 lines (44 loc) · 7.95 KB
/
Copy pathupdates.yaml
File metadata and controls
71 lines (44 loc) · 7.95 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
databases:
postgres-sqlite-best-practices: |-
Fetch PostgreSQL and SQLite primary documentation for source material only; do not include raw sitefetch output in the skill.
Run `sitefetch https://www.postgresql.org/docs/current/ -m '/docs/current/ddl-constraints.html' -m '/docs/current/datatype.html' -m '/docs/current/indexes.html' -m '/docs/current/indexes-multicolumn.html' -m '/docs/current/using-explain.html' -m '/docs/current/planner-stats.html' -m '/docs/current/sql-altertable.html' -m '/docs/current/sql-createindex.html' -m '/docs/current/tutorial-transactions.html' -m '/docs/current/explicit-locking.html' -m '/docs/current/transaction-iso.html' -m '/docs/current/ddl-priv.html' -m '/docs/current/ddl-rowsecurity.html' -m '/docs/current/routine-vacuuming.html' -m '/docs/current/monitoring.html' -m '/docs/current/datatype-json.html'` to fetch the PostgreSQL docs used by the skill.
Run `sitefetch https://sqlite.org/docs.html -m '/datatype3.html' -m '/foreignkeys.html' -m '/queryplanner.html' -m '/optoverview.html' -m '/eqp.html' -m '/lang_analyze.html' -m '/lang_altertable.html' -m '/isolation.html' -m '/wal.html' -m '/pragma.html' -m '/backup.html' -m '/lang_vacuum.html'` to fetch the SQLite docs used by the skill.
Rework the fetched material into concise, task-oriented guidance. Keep `databases/postgres-sqlite-best-practices/SKILL.md` as a routing signpost with the must-run final checklist, and keep task-specific details in `references/`. Prefer portable guidance first, then clearly label Postgres-only and SQLite-only behavior.
frontend:
motion-react: |-
Run `sitefetch https://motion.dev/docs/react -m '/docs/react-*' --link-source-selector '[id="learn-mobile-sidebar"]'` to fetch the React docs for source material only; do not include raw sitefetch output in the skill.
Rework the fetched documentation into concise, task-oriented guidance. Keep `frontend/motion-react/SKILL.md` focused on Motion primitives only, and put setup, migrations, accessibility, bundle-size, Motion+, integrations, and other non-primitive pages into individual guide files that closely match the source pages. Include a signpost in `SKILL.md` pointing to those guide files.
lingui: |-
Run `sitefetch https://lingui.dev/tutorials/javascript -m '/ref/macro' -m '/ref/vite-plugin' -m '/ref/conf' -m '/guides/dynamic-loading-catalogs'` to fetch the source docs for the vanilla JavaScript tutorial, core macro reference, Vite plugin reference, Lingui configuration reference, and dynamic catalog loading guide.
Rework the fetched material into concise, task-oriented guidance for using Lingui with `@lingui/core` and the Core Macro API. Do not include renderer-specific APIs or guidance for JSX frameworks such as React or Solid unless the user specifically asks for them.
lisse: |-
Run `pnpx degit JaceThings/Lisse.wiki.git /tmp/lisse-wiki` to fetch the Lisse wiki docs for source material only; do not include raw wiki output in the skill.
Rework the fetched documentation into concise, task-oriented greenfield React SPA guidance. Keep `frontend/lisse/SKILL.md` as a routing signpost and put detailed material in `references/`. Omit non-React renderer guidance, SSR/server/edge guidance, Tailwind-specific guidance, and migration guidance. Keep `@lisse/core`, performance, and limitations guidance only when it supports client-side React SPA use.
pragmatic-drag-and-drop: |-
Run `sitefetch https://atlassian.design/components/pragmatic-drag-and-drop/core-package -m '**/core-package/**' -o /tmp/pragmatic-dnd-core.txt` against the core-package docs for source material. If Atlassian's readable extraction is sparse for some pages or the crawl becomes too broad, fetch the same core-package pages explicitly with `--limit 0` and document which source file was used.
Rework the fetched documentation into concise, task-oriented guidance. Keep `frontend/pragmatic-drag-and-drop/SKILL.md` focused on operational decisions and common implementation/debugging patterns, and keep API details in `frontend/pragmatic-drag-and-drop/references/core-api.md`.
react-router: |-
Run `sitefetch https://reactrouter.com/start/data/installation -m '/start/data/**'` to fetch React Router source docs for client-owned route object configuration.
If a fetched page says to refer to another React Router start page, fetch that referenced page too. Use only guidance that applies to route objects, navigation, pending UI, and testing.
Rework the fetched documentation into concise, task-oriented React Router guidance. Keep `frontend/react-router/SKILL.md` focused on setup, lazy route components, route objects, typed route builders, URL/search-param conventions, shareable page state, navigation safety, pending UI, TanStack Query server state, and testing. Use fetched documentation as source material only; keep raw sitefetch output out of the skill.
react-virtual: |-
Run `sitefetch https://tanstack.com/virtual/latest/docs/introduction -m '/virtual/latest/docs/**' -e '**/examples/**'` to fetch the full package docs for source material only; do not include raw sitefetch output in the skill.
Run `pnpx degit TanStack/virtual/examples/react frontend/react-virtual/examples` to refresh the bundled React examples.
Rework the fetched documentation into concise, task-oriented reference files and update `frontend/react-virtual/SKILL.md` with the most useful operational knowledge. Keep setup guidance in `references/setup.md`, keep the skill description sharp and concise, and use `SKILL.md` for common use cases and routing to reference files and examples.
typescript:
ai-sdk: |-
Run `sitefetch https://ai-sdk.dev/docs/migration-guides/migration-guide-7-0 --limit 0` to fetch the AI SDK v7 migration documentation for source material only; do not include raw sitefetch output in the skill.
Rework the fetched documentation into concise greenfield AI SDK v7 guidance for agents that already understand AI SDK concepts. Keep all guidance in `typescript/ai-sdk/SKILL.md`, including current API names, package notes, result semantics, and provider-specific details.
Do not turn the skill into a step-by-step migration guide or v6-to-v7 comparison document. Assume v6 was never used, and teach v7 as the baseline for new projects.
cmd-ts: |-
Run `sitefetch https://cmd-ts.vercel.app/ -o /tmp/cmd-ts-docs.json` to fetch the full cmd-ts docs for source material only; do not include raw sitefetch output in the skill.
Cross-check API names and options against the published `cmd-ts` TypeScript declarations with `npm view cmd-ts version` and `npm pack cmd-ts`, because the docs pages may omit some heading text or newer exported helpers.
Rework the fetched documentation into concise, task-oriented guidance for building typed TypeScript CLIs with cmd-ts. Keep `typescript/cmd-ts/SKILL.md` focused on workflow, design decisions, and common implementation patterns, and keep parser/type/runtime API details in `typescript/cmd-ts/references/api.md`.
protocols:
agent-client-protocol: |-
Run `sitefetch https://agentclientprotocol.com/protocol/overview -m '/protocol/**'` to fetch the full protocol docs, then update `protocols/agent-client-protocol/SKILL.md` and its bundled references from the fetched documentation.
Prominently include this instruction: for projects using `@agentclientprotocol/sdk` from npm, check the TypeScript SDK changelog at https://raw.githubusercontent.com/agentclientprotocol/typescript-sdk/refs/heads/main/CHANGELOG.md when updating or validating SDK-specific guidance.
workflows:
sem: |-
Run `sitefetch https://ataraxy-labs.github.io/sem/details.html --limit 0` to fetch the sem commands and details page for source material.
Update `workflows/sem/SKILL.md` with concise, task-oriented guidance for all documented user-facing commands except setup/unsetup, including semantic diffs, impact analysis, blame, entity history logs, entity listing, and AI context generation. Do not include setup/unsetup commands, benchmark information, or machine-readable output guidance.