validate

Verify changes against acceptance criteria and repo quality gates to produce a pass/fail verdict.

3|Updated Jun 17, 2026
One-click install
npx skills add https://github.com/ronaknnathani/relay --skill validate-ronaknnathani
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: validate
Source: https://github.com/ronaknnathani/relay/tree/main/skills/validate
Command: npx skills add https://github.com/ronaknnathani/relay --skill validate-ronaknnathani

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Before shipping a change, you need an objective go/no-go decision: does the code actually meet its acceptance criteria, and do all of the repository's own quality gates pass? This Skill removes guesswork by running the repo's lint, typecheck, test, build, and audit commands and checking every acceptance criterion with concrete evidence, then reporting a clear PASS or FAIL verdict. ## Core Features & Use Cases - Acceptance criteria verification: Reads success criteria from clarify/plan artifacts and requires a named test, command, or observation as evidence for each one. - Repo-native quality gates: Discovers the repository's actual lint, typecheck, unit test, build, integration, e2e, security, and bundle-size commands instead of substituting generic ones. - Actionable failure reports: On FAIL, lists every failure with the exact command, error output, and file/line location so a fixer skill can act on it in one pass. - Use Case: After implementing a new API validation feature, run this Skill to confirm all tests pass, the build succeeds, and each planned acceptance criterion is demonstrably met before opening the pull request. ## Quick Start Ask the agent to validate the current change against its acceptance criteria and the repo's quality gates and report a pass or fail verdict.

Frequently Asked Questions about validate

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I validate a code change before shipping?▼

Run the repo's own quality gates in order — lint, typecheck, unit tests, build, integration, e2e, security audit, and bundle-size — then check each acceptance criterion with a named test or observation. The verdict is PASS only when all gates are green and every criterion has evidence.

What is the difference between code review and validation?▼

Validation produces a pass/fail verdict against acceptance criteria and quality gates, while code review critiques code quality and design. Use review for feedback on the code itself and validate for the go/no-go shipping decision.

Can validation fix the failures it finds?▼

No, validation only reports failures and never edits code. Each failure is reported with the exact command, error output, and location, then routed to a fixer skill such as pr-fix or implement.

What happens if a repo does not define a quality gate?▼

A gate the repo does not define is recorded as not-applicable rather than faked or substituted with a generic command. Gate commands are always discovered from the repo's own scripts, manifest, or CI configuration.

Why is disabling a failing test a red flag in validation?▼

Disabling a test, skipping a spec, or loosening a lint rule converts a real failure into a hidden one. Validation treats this as a hard red flag and requires reporting the failure with evidence instead of masking it to go green.