agentleFS
oaslanankaoaslananka/kicad-mcp-pro/AGENTS.md

kicad-mcp-pro

AGENTS.md169 starsChanged 18 days agoMITView on GitHub

Use this file as a router, not as a duplicate engineering manual. Follow the canonical sources below and keep changes evidence-driven. Nested operational boundaries: - src/kicadmcp/AGENTS.md — server architecture, KiCad adapter seam, path/subprocess safety, typed errors, and real-state rules. - src/kicadmcp/tools/AGENTS.md — public MCP tools, profiles, operating modes, annotations, generated contracts, and manufacturing gates. - .github/AGENTS.md — CI, required checks, workflow security, release, provenance, and publishing. - integrations/AGENTS.md — agent-runtime config templates, installers, permission defaults, and integration drift. - src-tauri/AGENTS.md…

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

6 rules. About 0 are common, 5 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈5

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use this file as a router, not as a duplicate engineering manual.

    line 3
  2. Distinctive:

    Prefer existing domain services and adapter seams.

    line 31
  3. Distinctive:

    Add or change tests before production behaviour changes, then run the smallest relevant suite and repository gates.

    line 32
  4. Distinctive:

    Treat generated files as outputs.

    line 33
  5. Distinctive:

    Do not claim native KiCad, manufacturing, or live-model success without the repository's required evidence.

    line 35
  6. Very specific:

    Use task verify for the normal local quality gate and task ci for the local CI equivalent when the pinned toolchain is available.

    line 38

≈ “Common” is an estimate: at least 1 in 5 of the 36,368 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

Repository Agent Router

Use this file as a router, not as a duplicate engineering manual. Follow the canonical sources below and keep changes evidence-driven.

Scope and precedence

  • This file applies repository-wide.
  • Nested AGENTS.md files add or narrow instructions for their subtree; the closest applicable file wins.
  • Nested instructions must not weaken repository security, architecture, compatibility, evidence, release, or human-approval constraints unless the underlying policy is intentionally changed in the same work.
  • Executable repository policy and tests remain authoritative if prose drifts.

Nested operational boundaries:

  • src/kicad_mcp/AGENTS.md — server architecture, KiCad adapter seam, path/subprocess safety, typed errors, and real-state rules.
  • src/kicad_mcp/tools/AGENTS.md — public MCP tools, profiles, operating modes, annotations, generated contracts, and manufacturing gates.
  • .github/AGENTS.md — CI, required checks, workflow security, release, provenance, and publishing.
  • integrations/AGENTS.md — agent-runtime config templates, installers, permission defaults, and integration drift.
  • src-tauri/AGENTS.md — Rust/Tauri desktop lifecycle, backend compatibility, loopback boundary, and packaging.
  • packages/AGENTS.md — npm wrapper, protocol schemas, fixtures, and KiCad companion package boundaries.

Start here

  • Architecture and dependency direction: ARCHITECTURE.md
  • Contribution and review contract: CONTRIBUTING.md
  • Security boundaries and reporting: SECURITY.md
  • Common local commands: Taskfile.yml
  • Agent profile/surface policy: docs/agents/progressive-disclosure.md
  • Tool-selection workflow: docs/tools-agent-workflow.md
  • Generated tool reference: docs/tools-reference.generated.md
  • KiCad compatibility contract: compatibility.yaml

Working rules

  1. Inspect the relevant architecture/service/adapter boundary before editing.
  2. Prefer existing domain services and adapter seams; do not bypass them with ad-hoc KiCad file rewriting.
  3. Add or change tests before production behaviour changes, then run the smallest relevant suite and repository gates.
  4. Treat generated files as outputs. Use their documented generator/check command instead of hand-editing them.
  5. Preserve explicit GitHub Actions permissions, SHA-pinned Actions, path guards, and human-only manufacturing controls.
  6. Do not claim native KiCad, manufacturing, or live-model success without the repository's required evidence.

Validation

Use task verify for the normal local quality gate and task ci for the local CI equivalent when the pinned toolchain is available. For change-scoped validation, scripts/hook_pre_push.py maps touched files to the required checks.

For agent/profile changes also run pnpm run profiles:check, pnpm run toolsets:check, and pnpm run tool-contracts:check. For workflow changes run pnpm run workflows:policy, pnpm run workflows:lint, and pnpm run workflows:security.

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.