rde-eval
Run a targeted local React Doctor Evals loop against an uncommitted rule change. Use after focused rule tests pass, while inspecting real open-source hits, or when rule-validate needs local false-positive evidence before pull request parity.
How do I install this agent skill?
npx skills add https://github.com/millionco/react-doctor --skill rde-evalIs this agent skill safe to install?
- Gen Agent Trust Hubpass
The rde-eval skill is a development tool for testing react-doctor lint rules against a corpus of repositories. It facilitates running evaluations locally or in a Vercel-hosted sandbox. The skill uses standard development practices, such as git for version control and pnpm for package management, and follows security best practices by instructing users to manage credentials via environment files. No malicious patterns were detected.
- Socketpass
No alerts
- Snykwarn
Risk: MEDIUM · 2 issues
What does this agent skill do?
Run a local rule evaluation
Use React Doctor Evals (RDE) for bounded local iteration. Use run-parity only after the change has a pushed pull request.
Prepare both checkouts
export REACT_DOCTOR_CHECKOUT=/absolute/path/to/react-doctor
export RDE_CHECKOUT=/absolute/path/to/react-doctor-evals
git -C "$RDE_CHECKOUT" pull --ff-only
ni -C "$RDE_CHECKOUT"
nr -C "$RDE_CHECKOUT" build
nr -C "$REACT_DOCTOR_CHECKOUT" build
The path: spec reads uncommitted React Doctor changes. Run RDE commands from the eval checkout.
Run a bounded sample
cd "$RDE_CHECKOUT"
node dist/cli.js run "path:$REACT_DOCTOR_CHECKOUT" --runner local --take 100
node dist/cli.js digest "path:$REACT_DOCTOR_CHECKOUT" --rule <rule-id>
node dist/cli.js digest "path:$REACT_DOCTOR_CHECKOUT" --json --rule <rule-id> > <artifact-directory>/hits.json
Increase --take only after tests and the first sample pass.
Inspect target-rule hits
For each hit, or a representative sample when counts are high:
- Open the pinned repository at the reported location.
- Compare the code with the rule contract.
- Classify the hit as true positive, false positive, or unsupported.
- Add a rule regression test for each false positive.
- Add confirmed false positives to the
fuzzregression corpus. - Rebuild and rerun the same sample.
Record repository count separately from project-root count. Do not treat error records as clean scans.
Report results
Report checkout revisions, target rule, repositories, project roots, diagnostics, inspected hits, fixed false positives, and the artifact path. State any setup error or skipped repository.
After local validation, return to rule-validate. That skill decides whether to invoke pull request parity.
How can the creator link this skill?
Add the canonical catalog link to the repository README so users can inspect current installs and available audits. The publishing guide covers the complete discovery path.
<a href="https://skillzs.dev/skills/millionco/react-doctor/rde-eval">View rde-eval on skillZs</a>