agentleFS
Sign inStart free
euxxeuxx/claude-skills-for-copilot/AGENTS.md

claude-skills-for-copilot

AGENTS.md4 starsChanged 7 months agoMITView on GitHub

After modifying code, run: See DEVELOPMENT.md for all available scripts. See RELEASE.md for the release process. Active upstream repos: Legacy source: https://github.com/openai/skills (local mirror at ~/projects/skills) was deprecated on 2026-06-22. Its final checked snapshot is 49f948f; retain it for attribution and historical comparisons, not as the active OpenAI update source. Last upstream check: 2026-11-02 Verified on 2026-11-02 against OpenAI Plugins 0722921d:

The file

We split this AGENTS.md into the rules it gives an agent

1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    After modifying code, run:

    line 5

≈ “Common” is an estimate: at least 1 in 5 of the 40,246 AGENTS.md files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
AGENTS.md

Project Conventions

After Making Changes

After modifying code, run:

  • npm run ci — syntax + lint + format + tests

See DEVELOPMENT.md for all available scripts.

See RELEASE.md for the release process.

<!-- END-SHARED -->

Upstream Adaptation Tracking

Active upstream repos:

  • Anthropic: https://github.com/anthropics/claude-plugins-official — local mirror at ~/projects/claude-plugins-official
  • OpenAI Plugins: https://github.com/openai/plugins — local mirror at ~/projects/openai-plugins

Legacy source: https://github.com/openai/skills (local mirror at ~/projects/skills) was deprecated on 2026-06-22. Its final checked snapshot is 49f948f; retain it for attribution and historical comparisons, not as the active OpenAI update source.

Last upstream check: 2026-11-02

  • Anthropic checked up to commit a7ba5b93
  • OpenAI Plugins initial catalog inventory at commit 0722921d; this establishes a tracking baseline, not a completed adaptation review of every plugin.

How to check for updates

  1. Sync local mirrors — cd ~/projects/claude-plugins-official && git pull --ff-only and the same for ~/projects/openai-plugins. Confirm git status -sb shows no drift before comparing.
  2. List commits since last check — git log --oneline <last-sha>..HEAD in each active mirror. Do not compare SHAs across the old OpenAI Skills and new OpenAI Plugins repositories.
  3. Filter to adapted skills — restrict the log with -- <paths> for plugins/folders that map to skills in this repo. Current mapping:
    • plugins/code-review/ → code-review
    • plugins/claude-md-management/ → conventions-improver
    • plugins/pr-review-toolkit/ → code-review, comment-analyzer, silent-failure-hunter, test-analyzer, type-design-analyzer, review-toolkit
    • plugins/code-simplifier/ → code-simplifier
    • plugins/feature-dev/ → feature-dev
    • plugins/frontend-design/ → frontend-design
    • plugins/mcp-server-dev/ → mcp-server-dev
    • OpenAI mappings and related candidates are listed below. A related candidate is not an automatic replacement for an existing skill.
  4. Inspect each candidate commit — read the diff (git show <sha> -- <path>) and decide whether it touches content we actually carry (our SKILL.md files are often condensed single-file versions, so deep-reference-only changes may be no-ops).
  5. Look for newly added skills — git log --diff-filter=A --name-only --pretty=format: <last-sha>..HEAD -- 'plugins/*/skills/*/SKILL.md' in both active mirrors. Also inspect renames and deletions. Evaluate candidate content, host/tool dependencies, and its own license before adapting; do not infer a successor from a similar name.
  6. Record the new SHAs and date above after the check completes, regardless of whether any sync was needed.

OpenAI Source Migration

Verified on 2026-11-02 against OpenAI Plugins 0722921d:

Local skillLegacy source under skills/.curated/Current tracking status
gh-address-commentsgh-address-comments/No current same-name skill. plugins/github/skills/gh-address-comments/ was deleted by 33bd9529 on 2026-08-26. Keep the legacy adaptation.
gh-fix-cigh-fix-ci/No current same-name skill. plugins/github/skills/gh-fix-ci/ was deleted by 33bd9529 on 2026-08-26. Keep the legacy adaptation.
security-threat-modelsecurity-threat-model/Related candidate: plugins/codex-security/skills/threat-model/ and plugins/codex-security/references/. The entrypoint depends on scan artifact storage, checkpoints, policy, and cache conventions; it is not a drop-in successor.
frontend-layoutfrontend-skill/plugins/build-web-apps/skills/frontend-skill/ was deleted by ca0c87f6 on 2026-04-23. Related candidates: plugins/build-web-apps/skills/frontend-app-builder/ and plugins/product-design/. They introduce Image Gen/browser workflows and plugin-specific dependencies; no direct replacement has been established.

Monitor the related candidate paths for changes. This initial inventory identified the existing skills' status and inspected the related entrypoints; the rest of the new catalog has not yet received a full Copilot-fit review. Keep the legacy skill source links until a replacement is actually adapted, and preserve original license attribution.

Skills evaluated and skipped

  • session-report (Anthropic, added 2026-04) — depends on reading ~/.claude/projects Claude Code transcripts; not portable to VS Code Copilot.
  • cardputer-buddy, m5-onboard (Anthropic cwc-makers, added 2026-05) — M5Stack ESP32 hardware provisioning over USB serial / BLE; not applicable to VS Code Copilot.
  • define-goal (OpenAI, added 2026-05) — depends on Codex get_goal / create_goal tools that have no Copilot equivalent.
  • hatch-pet (OpenAI, added 2026-05) — Codex pet sprite-atlas workflow built on the $imagegen system skill; not applicable to VS Code Copilot.
  • project-artifact (Anthropic, added 2026-06) — depends on Claude Code's built-in Artifact tool, claude.ai/code/artifact/... publishing, and ${CLAUDE_PLUGIN_DATA} config state; not portable to VS Code Copilot.
  • claude-security (Anthropic, added 2026-07) — depends on Claude Code's Workflow, Agent, and AskUserQuestion tools plus plugin-relative scripts and multi-agent orchestration; not portable as a standalone VS Code Copilot skill.
  • receipts (Anthropic, added 2026-07) — mines Claude Code session transcripts under ~/.claude/projects and reports Claude Code-specific usage; not portable to VS Code Copilot.
  • solo (Anthropic math-proof, added 2026-09) — a pure research-mathematics proof workflow; technically adaptable, but outside this extension's current software-development scope.
  • siege (Anthropic math-proof, added 2026-09) — a long-running mathematics protocol requiring plugin-scoped judge/worker agents, a bundled Python ledger, and Claude Code runtime settings; not suitable for a standalone Copilot skill port.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

Reviews can't be read right now.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free