agentleFS
Sign inStart free
docxologydocxology/template/rules/AGENTS.md

template / rules

AGENTS.md20 starsChanged 2 months agoApache-2.0View on GitHub

This document tells AI agents how to discover, interpret, and enforce rules defined in this directory. When entering any project or manuscript context, check for a rules/ sibling or parent directory. Load rules.yaml from the most-specific matching rule set first, then fall back to broader scopes. - Treat as guidelines and preferences. - Apply judgment: follow the spirit, not the letter. - When generating text, code, or reviews, surface any deviation as a suggestion, not a blocker. - Treat as…

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

7 rules. About 0 are common, 5 are distinctive, and 2 are 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.

  • 2

    Very specific

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

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

    When entering any project or manuscript context, check for a rules/ sibling or parent directory.

    line 10
  2. Distinctive:

    Treat as guidelines and preferences.

    line 25
  3. Distinctive:

    When generating text, code, or reviews, surface any deviation as a suggestion, not a blocker.

    line 27
  4. Distinctive:

    Treat as hard constraints.

    line 32
  5. Distinctive:

    If a strong rule is violated, stop and report the violation with the path of the offending file and the constraint that was broken.

    line 34
  6. Distinctive:

    Do not silently skip strong-rule checks.

    line 36
  7. Very specific:

    Update rules.yaml metadata (type, description, version, scope).

    line 52

≈ “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

AGENTS — Rules Directory

This document tells AI agents how to discover, interpret, and enforce rules defined in this directory.

Discovery

When entering any project or manuscript context, check for a rules/ sibling or parent directory. Load rules.yaml from the most-specific matching rule set first, then fall back to broader scopes.

rules/<project-slug>/rules.yaml   # most specific
rules/templates/<template>/rules.yaml  # fallback exemplar

Interpreting Rule Kinds

Soft rules (soft/)

  • Treat as guidelines and preferences.
  • Apply judgment: follow the spirit, not the letter.
  • When generating text, code, or reviews, surface any deviation as a suggestion, not a blocker.

Strong rules (strong/)

  • Treat as hard constraints.
  • Validate them before marking a task complete.
  • If a strong rule is violated, stop and report the violation with the path of the offending file and the constraint that was broken.
  • Do not silently skip strong-rule checks.

Enforcement Workflow

  1. Load rules.yaml → identify rule_kinds present.
  2. For each strong/ schema, run the described validation.
  3. For each soft/ guideline, apply as a prompt constraint during generation.
  4. Report results: pass / warn / fail per rule.

Adding New Rules

  1. Copy the appropriate template from rules/templates/.
  2. Update rules.yaml metadata (type, description, version, scope).
  3. Edit soft guidelines in soft/.
  4. Edit strong schemas in strong/.
  5. Bump version if any strong constraint changes.

Key Files

FilePurpose
rules.yamlMachine-readable rule set metadata
soft/*.mdHuman/agent-readable guideline prose
strong/*.yamlMachine-enforceable constraint schemas
README.mdHuman-readable rule set overview
AGENTS.mdThis file — agent-facing instructions

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