agentleFS
Sign inStart free
OpenSourcxOpenSourcx/papercliped/AGENTS.md

papercliped

AGENTS.md12 starsChanged 38 days agoMITView on GitHub

Papercliped connects AI apps (Claude, ChatGPT, Codex, any MCP client) to a Paperclip instance. This file is read by Claude Code (CLAUDE.md) and Codex (AGENTS.md); keep the two identical. Use the release skill (.claude/skills/release/SKILL.md, .agents/skills/release/SKILL.md) before every push to main: pick Major.Minor.Patch, add the user-facing entry under ## Unreleased in CHANGELOG.md, run node scripts/version.mjs <major|minor|patch> "<title>", and write the commit description as the release note. CI publishes to npm, tags the commit and creates the GitHub release.

  • Where files are headingNew AGENTS.md files have more rules about keeping secrets out of the code, and fewer about code style.

    Of the AGENTS.md files we read, the 31,700 first seen in the last six months, against the older ones: the kind of rule that gained most share, and the kind that lost most. About files like this in general, not this one, and an estimate, since a kind is a pattern in the words.

  • In this repository4 other files it gives its agents.

The file

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

3 rules. About 0 are common, 1 is distinctive, and 2 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.

  • 2

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  • could be automatic: a setting could enforce it
  1. Very specific:

    Use the release skill (.claude/skills/release/SKILL.md, .agents/skills/release/SKILL.md) before every push to main: pick Major.Minor.Patch, …

    line 7
  2. Very specific:

    Never commit secrets. Never weaken a security check, test or the website's CSP (script-src 'self'.

    line 32
  3. Distinctive:

    Don't put AI model names in commits, code or docs.

    line 37

≈ “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

Working on Papercliped

Papercliped connects AI apps (Claude, ChatGPT, Codex, any MCP client) to a Paperclip instance. This file is read by Claude Code (CLAUDE.md) and Codex (AGENTS.md); keep the two identical.

Every push is a release

Use the release skill (.claude/skills/release/SKILL.md, .agents/skills/release/SKILL.md) before every push to main: pick Major.Minor.Patch, add the user-facing entry under ## Unreleased in CHANGELOG.md, run node scripts/version.mjs <major|minor|patch> "<title>", and write the commit description as the release note. CI publishes to npm, tags the commit and creates the GitHub release.

Map

PathWhat
src/The bridge: MCP + HTTP server, OAuth (src/oauth/), tools (src/tools.ts), manage API (src/manage/), public API, website serving (src/web/)
plugin/Papercliped plugin for Paperclip (own package; Node 24.11+)
web/Website and docs site (React + Tailwind, Vite)
site/docs/User docs in Markdown (rendered by the website and the bridge)
migrations/Postgres schema; regenerate docs/supabase-schema.sql with npm run schema:sql
.claude-plugin/, .codex-plugin/, .agents/plugins/, .mcp.json, skills/, commands/The Claude Code and Codex plugins and marketplaces

Checks

npm run typecheck && npm test                     # bridge (Postgres tests need TEST_DATABASE_URL)
npm --prefix plugin run typecheck && npm --prefix plugin test && npm --prefix plugin run build
npm --prefix web test && npm --prefix web run build && npm --prefix web run check:routes
node scripts/check-packages.mjs                    # after building root and plugin
node scripts/version.mjs check

Rules

  • Never commit secrets. Never weaken a security check, test or the website's CSP (script-src 'self'; style-src 'self', no inline code, no third-party loads, no analytics) to make something pass.
  • Access levels are exactly two: Read only (paperclip:read) and Full control (paperclip:control, every tool).
  • The public stats API is aggregate-only; it never exposes identities.
  • Community links stay empty until the real accounts exist; don't invent handles.
  • Domain: papercliped.co (site), docs.papercliped.co (docs at /<page>), mcp.papercliped.co (connector /mcp).
  • Don't put AI model names in commits, code or docs.

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.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

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