agentleFS
WrosinskiWrosinski/heddle/AGENTS.md

heddle

AGENTS.md0 starsChanged 55 days agoMITView on GitHub

This repository's feature workflow runs through the heddle CLI. To enter a feature-work session: Engineering preferences are governed by docs/workflow/engineering-principles.md. Feature execution records stay local under the paths declared in release/public-repository.toml. Never stage or force-add those records. Keep proposals in the ignored proposal directory; publish maintained product documentation separately from implementation history. Follow the repository policy in docs/workflow/local-records.md even when packaged guidance permits optional record commits. Tests use synthetic fixtures, never local run evidence. Test selection, execution authority, design and…

The file

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

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

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    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:

    Follow the next_actions it returns.

    line 8
  2. Very specific:

    Run heddle kickoff only when orient routes there.

    line 9
  3. Very specific:

    Keep proposals in the ignored proposal directory.

    line 17
  4. Distinctive:

    Test selection, execution authority, design and maintenance are governed by the testing strategy.

    line 22
  5. Distinctive:

    Run narrow relevant hermetic tests, configured lint and applicable pre-commit checks as part of the requested work.

    line 29
  6. Distinctive:

    Keep behavior changes and mechanical changes (formatting, renames, moves) in separate commits.

    line 42

≈ “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
<!-- heddle:begin session-entry -->

This repository's feature workflow runs through the heddle CLI.

To enter a feature-work session:

  1. Run heddle orient.
  2. Follow the next_actions it returns.
  3. Run heddle kickoff only when orient routes there.
<!-- heddle:end session-entry -->

Engineering preferences are governed by docs/workflow/engineering-principles.md.

Feature execution records stay local under the paths declared in release/public-repository.toml. Never stage or force-add those records. Keep proposals in the ignored proposal directory; publish maintained product documentation separately from implementation history. Follow the repository policy in docs/workflow/local-records.md even when packaged guidance permits optional record commits. Tests use synthetic fixtures, never local run evidence.

Test selection, execution authority, design and maintenance are governed by the testing strategy. Consult its current capability notes and selection map before choosing commands. Repository pytest policy is active: plain pytest selects the fast band, while formal proof uses .venv/bin/python -m tests.proof_runner <exact-targets>.

Run narrow relevant hermetic tests, configured lint and applicable pre-commit checks as part of the requested work. Fix failures caused by the change and rerun affected checks without asking at each step. Broad suites and both local full-workflow e2e and real-provider tests require explicit execution authority for that scope; existing applicable grants persist. Prerequisites and inherited plan commands are not permission. Preserve required unrun proof as pending and inspect nested launches before execution. Use --test-band e2e --allow-e2e or --test-band live --allow-live --allow-e2e only for an exactly authorized file or node selection.

Committing: use scripts/git-commit-retry.sh <git-commit args> — when pre-commit hooks reformat staged files it re-stages exactly those files and retries once; any other failure propagates unchanged. Keep behavior changes and mechanical changes (formatting, renames, moves) in separate commits.

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.