public-docs
Create and maintain public, user-facing product documentation for this repo's `packages/docs` app. Use when asked to write, revise, or review docs pages (tutorial/concept/reference/troubleshooting), improve docs information architecture, fix docs frontmatter/linking issues, or align docs quality with Vercel/Sentry-style standards.
Works with
--- name: public-docs description: Create and maintain public, user-facing product documentation for this repo's `packages/docs` app. Use when asked to write, revise, or review docs pages (tutorial/concept/reference/troubleshooting), improve docs information architecture, fix docs frontmatter/linking issues, or align docs quality with Vercel/Sentry-style standards. license: Apache-2.0 --- # Public Docs Use this workflow to produce docs that are crisp, task-oriented, and trustworthy for external users. Load only the references needed for the request: | Task | Read | |------|------| | Understand docs app layout and constraints | `references/api-surface.md` | | Draft or revise common docs pages | `references/common-use-cases.md` | | Diagnose docs quality/build failures | `references/troubleshooting-workarounds.md` | ## Core invariants 1. Keep docs audience-facing: explain user outcomes, not internal implementation chatter. 2. Match the local docs contract in `packages/docs/src/content/docs/contribute/documentation-guidelines.md`. 3. For new or substantially edited pages, require frontmatter fields: `type`, `summary`, `prerequisites`, `related`. 4. Include concrete verification steps for tutorial/troubleshooting content. 5. End each page with an explicit "next step" path through internal links. 6. Prefer minimal runnable examples over long conceptual exposition. 7. Keep prose present around lists; avoid heading-plus-bullets-only sections. ## Workflow ### 1. Classify the request Select one primary path: - New page authoring - Existing page refactor/clarification - Information architecture update (sidebar/redirects/navigation) - Docs quality pass (accuracy, scannability, consistency) - Troubleshooting/build-fix for docs app ### 2. Gather local context 1. Read `packages/docs/src/content/docs/contribute/documentation-guidelines.md`. 2. Read neighboring docs in the same section (`start-here`, `concepts`, `extend`, `operate`, `reference`, `contribute`). 3. If navigation is involved, read `packages/docs/astro.config.mjs` (`sidebar` and `redirects`). 4. If public interfaces are involved, verify the exported TypeScript surface and its code comments directly. ### 3. Author or edit with the right page shape 1. Pick doc `type` (`conceptual`, `tutorial`, `reference`, `troubleshooting`) from intent. 2. Follow depth and section guidance from documentation guidelines. 3. Keep wording direct and plain; remove filler or internal-only caveats. 4. Use titled code fences for file snippets. 5. Prefer internal links that advance the reader to the next concrete action. ### 4. Validate changes Run the narrowest meaningful checks: ```bash pnpm docs:check ``` When structure, sidebar, or redirects change, also run: ```bash pnpm docs:build ``` ### 5. Report completion Return: 1. Changed files and what each change improves for readers 2. Validation commands executed and outcome 3. Residual risks (for example, missing screenshots, unverified env-specific setup)
More Writing & Documentation skills
paper-context-resolver
lllllllama/rigorpilot-skills
Rigor Paper Context helper for README-first deep learning repo reproduction. Use only when the README and repository files leave a narrow reproduction-critical gap and the task is to resolve a specific paper detail such as dataset split, preprocessing, evaluation protocol, checkpoint mapping, or runtime assumption from primary paper sources while recording conflicts. Do not use for general paper summary, repo scanning, environment setup, command execution, title-only paper lookup, or replacing README guidance by default.
repo-intake-and-plan
lllllllama/rigorpilot-skills
Rigor Intake helper for README-first deep learning repo reproduction. Use when the task is specifically to scan a repository, read the README and common project files, extract documented commands, classify inference, evaluation, and training candidates, and return the smallest trustworthy reproduction plan to the main orchestrator. Do not use for environment setup, asset download, command execution, final reporting, paper lookup, or end-to-end orchestration.
minimal-run-and-audit
lllllllama/rigorpilot-skills
Rigor Run skill for README-first deep learning repo reproduction. Use when the task is specifically to capture or normalize evidence from the selected smoke test or documented inference or evaluation command and write standardized `repro_outputs/` files, including patch notes when repository files changed. Do not use for training execution, initial repo intake, generic environment setup, paper lookup, target selection, hidden scientific-meaning changes, or end-to-end orchestration by itself.

