CtrlK
BlogDocsLog inGet started
Tessl Logo

putio/frontend-docs

Structure and rewrite docs for frontend repositories, especially README.md, CONTRIBUTING.md, SECURITY.md, and other top-level docs. Use when creating or reorganizing frontend repo docs, clarifying user vs contributor guidance, reducing doc sprawl, or fixing stale commands, paths, and links in top-level docs.

95

Quality

95%

Does it follow best practices?

Impact

Pending

No eval scenarios have been run

SecuritybySnyk

Passed

No known issues

Overview
Quality
Evals
Security
Files

SKILL.md

name:
putio-frontend-docs
description:
Structure and rewrite docs for frontend repositories, especially README.md, CONTRIBUTING.md, SECURITY.md, and other top-level docs. Use when creating or reorganizing frontend repo docs, clarifying user vs contributor guidance, reducing doc sprawl, or fixing stale commands, paths, and links in top-level docs.

Frontend Docs

Shape frontend repo docs around a clear split between user-facing docs and contributor-facing docs. Treat structure, labels, and ordering as part of the interface, not just as neutral file plumbing.

Workflow

  1. Inspect the repository before drafting.
  2. Identify the project type, user install and usage flow, contributor setup flow, any GitHub collaboration templates already in use, and the best existing docs to link.
  3. Read references/readme-guideline.md before picking a final shape.
  4. Start from references/contributing-template.md when creating or reshaping CONTRIBUTING.md.
  5. Start from references/security-template.md when creating or reshaping SECURITY.md.
  6. Put each concern in its canonical home: user flow in README.md, contributor workflow in CONTRIBUTING.md, security reporting in SECURITY.md, and detailed review prompts in GitHub templates.
  7. Ensure the repo has CONTRIBUTING.md, LICENSE, and SECURITY.md; keep SECURITY.md private-first and use devs@put.io for security contact.
  8. Push deep implementation detail into linked docs when it starts to bloat the top-level docs.
  9. When a repo uses AGENTS.md, keep CLAUDE.md beside it as a symlink to AGENTS.md instead of maintaining a second authored guidance file.
  10. In checked-in docs, use repo-relative Markdown links for local files. Reserve absolute filesystem paths for chat/UI file references, not versioned docs.
  11. Verify that every claimed command, path, email address, doc link, badge target, and GitHub template path exists.
  12. If any command, path, link, badge target, or template reference is broken, fix the doc and re-verify before stopping.

Concrete checks:

rg -n "README|CONTRIBUTING|SECURITY|AGENTS|CLAUDE|docs/|pull_request_template|ISSUE_TEMPLATE" README.md CONTRIBUTING.md SECURITY.md AGENTS.md CLAUDE.md docs/ .github/
test -e README.md && test -e CONTRIBUTING.md
test ! -e AGENTS.md || { test -L CLAUDE.md && test "$(readlink CLAUDE.md)" = "AGENTS.md"; }

Guardrails

  • Do not guess commands, environment variables, or deployment behavior.
  • Prefer one canonical location per fact. Point to docs or templates instead of copying long checklists between them.
  • Do not hardcode volatile metrics such as test counts or coverage numbers.
  • Do not add generic filler sections that say nothing specific about the repo.
  • Do not cite or link unrelated external repos in generated docs unless the user explicitly asks for that.
  • Do not leak chat-only absolute filesystem links such as /Users/..., file://..., or vscode://... into checked-in docs.
  • Do not assume the repo provides a link-checking script or any repo-specific docs helper unless it actually exists.
  • Never include user PII (names, emails, usernames, IPs, etc.) in docs, references, or examples. Mask or normalize any PII to generic placeholders (e.g., user@example.com, your-username).
  • Never mention third-party applications by name in docs or references. Refer to them collectively as "ecosystem apps".

SKILL.md

tile.json