Research a rule

SkillDev tools

Lets your agent research and define a precise React Doctor rule contract before writing the rule.

Available today. Use it from your connected AI after setup.

Connect ahel once, and every AI you use reads what you have installed.

Then ask your AI: use the Research a rule skill

About this capability

Define a precise React Doctor rule contract before implementation. Use when validating a rule idea, collecting official or open-source evidence, identifying false-positive traps, choosing detector precision, or setting first-version boundaries.

What this skill tells your AI

The instructions your AI receives, as published by millionco/react-doctor in .agents/skills/rule-research/SKILL.md and read by ahel’s review.

Produce a rule contract that rule-writing can implement without guessing.

Define the behavior

Resolve only questions that affect correctness:

  • Which code pattern should report?
  • Which runtime behavior makes it harmful?
  • Which similar code must stay quiet?
  • Does detection need syntax, scope, or path analysis?
  • Which imported, dynamic, type-driven, or interprocedural cases stay out of scope?

If the user requested implementation, make the contract concise and continue.

Collect evidence

  1. Define the rule in one sentence: This rule catches <pattern> that causes <problem>.

  2. Explain the runtime reason.

  3. Inspect nearby rules, tests, utilities, and the generated registry.

  4. Use truffler before proposing a new detector or helper:

    bunx @rayhanadev/truffler "<symbol-or-behavior>" \
      packages/oxlint-plugin-react-doctor/src/plugin \
      --kind function,interface,type,constant --limit 20
    
  5. Gather official documentation, implementation notes, related linter behavior, and open-source examples.

  6. Separate strong positives, adjacent patterns, valid traps, and unsupported cases.

  7. Choose syntax-only, scope-aware, or path-aware detection.

Use rde-eval when a bounded open-source sample could change the contract. Leave final pull request parity to rule-validate.

Write the contract

Return:

Rule definition:
<pattern and specific problem>

Runtime reason:
<short explanation>

Detector precision:
<syntax-only, scope-aware, or path-aware>

Evidence:

- <source and implication>

Strong positives:

- <reportable examples>

False-positive traps:

- <valid examples>

In scope:

- <supported cases>

Out of scope:

- <explicit boundaries>

Test seeds:

- <invalid and valid fixtures>

Open questions:

- <correctness blockers only>

Treat false positives as correctness bugs. Keep the diagnostic narrower than or equal to the proven behavior. Split adjacent ideas into separate rules.

Signals

GitHub stars
15k
Forks
479
Last commit
Sep 2026
Advanced
Catalog kind
skill
Gateway key
rule-research
Source
github.com/millionco/react-doctor