agentleFS
Sign inStart free
2ykwang2ykwang/cview/AGENTS.md

cview

AGENTS.md19 starsChanged 3 months agoMITView on GitHub

Repository instructions for coding agents. README.md is the human-facing product and contributor guide; keep agent-specific guidance here. Use the Node version in .node-version. In a fresh checkout, install the locked dependencies and Chromium once: Use the smallest relevant test while iterating, then run the appropriate gate: Run npm run check before handing off any source or configuration change. Run npm run check:all when a change can cross the server/browser boundary or affect navigation, themes, rendering, export, capture, or build configuration.…

  • 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 repository3 other files it gives its agents.

The file

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

7 rules. About 0 are common, 4 are distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈4

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 3

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Use the Node version in .node-version.

    line 8
  2. Distinctive:

    Use the smallest relevant test while iterating, then run the appropriate gate:

    line 15
  3. Very specific:

    Run npm run check before handing off any source or configuration change.

    line 23
  4. Very specific:

    Keep ~/.claude read-only and write derived indexes to the external cache.

    line 45
  5. Distinctive:

    Never copy real session files into tests.

    line 46
  6. Distinctive:

    Keep cold filesystem scans bounded.

    line 49
  7. Distinctive:

    Keep heavy export/capture dependencies behind the existing lazy boundary.

    line 55

≈ “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

AGENTS.md

Repository instructions for coding agents. README.md is the human-facing product and contributor guide; keep agent-specific guidance here.

Bootstrap and verification

Use the Node version in .node-version. In a fresh checkout, install the locked dependencies and Chromium once:

npm run bootstrap

Use the smallest relevant test while iterating, then run the appropriate gate:

npm test -- path/to/file.test.ts  # focused Vitest run
npm run check                     # types, lint, design lint, Vitest, build
npm run check:all                 # check plus production Playwright flows

Run npm run check before handing off any source or configuration change. Run npm run check:all when a change can cross the server/browser boundary or affect navigation, themes, rendering, export, capture, or build configuration. For DESIGN.md-only work, npm run lint:design is the focused gate.

Where behavior belongs

  • src/server.ts composes the app; filesystem, search, security, and HTTP/SSE behavior belongs in focused modules under src/server/.
  • src/shared/contracts.ts parses and normalizes JSONL records and API responses.
  • Frontend transport belongs in api/, pure conversation behavior in domain/, browser lifecycles in hooks/, and screen/display work in pages/ and components/.
  • Live and standalone export behavior belongs in src/frontend/src/export/.
  • For UI or export presentation changes, read DESIGN.md. Semantic colors live in styles/theme.css, shared conversation rules in styles/conversation.css, and typed React references in styles/tokens.ts.

Contracts to preserve

  • Parse Claude records through the shared contracts and keep unknown record kinds visible instead of silently discarding them.
  • Keep ~/.claude read-only and write derived indexes to the external cache. Never copy real session files into tests; use small JSONL files with invented data.
  • Bind to loopback, reject non-local API clients/origins, and validate every project, session, and agent path segment before filesystem access.
  • Keep cold filesystem scans bounded; do not fan out unbounded reads across all sessions.
  • Preserve public CLI names, environment variables, API behavior, stored-data compatibility, message ordering, and stable message IDs unless requested.
  • Standalone HTML must remain offline-capable. Re-export must not nest controls or create multiple toolbars, and capture must work in both themes.
  • Keep heavy export/capture dependencies behind the existing lazy boundary.

Change discipline

Preserve unrelated work in a dirty tree. Prefer tests at the closest useful boundary and add Playwright coverage only for behavior that requires a real browser or crosses runtimes. Do not edit or commit generated build/, dist/, Playwright artifacts, or package tarballs. When user-facing documentation changes, regenerate tracked README screenshots with bash scripts/capture-demo.sh instead of editing them by hand.

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