enforce-rules-format
Skill15 starsChanged 5 months agoMITView on GitHub
Validate that all rules have proper enforcement classification (enforce/check/guidance)
In this repository24 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- examples/CLAUDE.mdCLAUDE.md
- src/CLAUDE.mdCLAUDE.md
- add-a-linterSkill
- audience-checkSkill
- audit-feedback-loopSkill
- code-qualitySkill
- deep-researchSkill
- dogfood-cliSkill
- generate-logoSkill
- hexagonal-architectureSkill
- landing-siteSkill
- pr-to-lint-ruleSkill
- review-docsSkill
- screenshotSkill
- ship-a-featureSkill
- verify-docs-findableSkill
- pr-to-lint-ruleSkill
- adopt-specSkill
- debug-my-harnessSkill
- edit-specSkill
- linter-docsSkill
- strengthenSkill
- test-harnessSkill
The file
We split this skill into the rules it gives an agent
9 rules. About 0 are common, 5 are distinctive, and 4 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.
4
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 13
Check which format the project uses:
- Very specific: line 15
v2 (spec-based): Look for CLAUDE.md.spec.ts or any *.spec.ts files.
- Very specific: line 17
v1 (hand-written): Look for CLAUDE.md, AGENTS.md, .cursorrules.
- Very specific: line 25
Do enforce() rules reference real linter rules?
- Distinctive: line 48
Check the project's linter configuration for matching rules
- Very specific: line 50
If no linter rule exists, suggest guidance() (v2) or Guidance only (v1)
- Distinctive: line 51
Ask the user: before making changes
- Distinctive: line 55
If the project uses v1 format, suggest migrating to v2 specs for type safety:
- Distinctive: line 61
Run the appropriate command:
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill 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.
Files
Validate that every rule in the project's instruction files has a proper enforcement classification, and fix any that are missing.
Instructions
Step 1: Detect Format
Check which format the project uses:
v2 (spec-based): Look for CLAUDE.md.spec.ts or any *.spec.ts files. If found, this is a v2 project — rules must use enforce(), check(), or guidance().
v1 (hand-written): Look for CLAUDE.md, AGENTS.md, .cursorrules. If found without a spec file, this is a v1 project — rules need **Enforced by:** or **Guidance only** annotations.
Step 2: Validate
For v2 specs:
The TypeScript type system already prevents unannotated rules — you can't create a rule without calling enforce(), check(), or guidance(). So focus on:
- Do
enforce()rules reference real linter rules? Runnpx vigiles compileto check. - Are there guidance rules that COULD be
enforce()? Check linter configs for matching rules. - Are there
check()assertions that could be delegated to a linter? Suggestenforce()instead.
npx vigiles compile
npx vigiles discoverFor v1 hand-written files:
Scan for ### headings. Each must have one of:
**Enforced by:** \linter/rule-name``**Guidance only** — reason<!-- vigiles-disable -->
Report missing annotations with a summary table.
Step 3: Fix Issues
For each issue found:
- Check the project's linter configuration for matching rules
- Suggest
enforce("linter/rule")(v2) or**Enforced by:** \linter/rule`` (v1) - If no linter rule exists, suggest
guidance()(v2) or**Guidance only**(v1) - Ask the user before making changes
Step 4: Suggest Migration
If the project uses v1 format, suggest migrating to v2 specs for type safety:
Your rules could benefit from type-safe specs. Run the
adopt-specskill to convert your CLAUDE.md to a typed .spec.ts file.
Step 5: Verify
Run the appropriate command:
# v2
npx vigiles compile && npx vigiles check
# validate
npx vigiles checkReport the validation result.
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
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.

