uinaf/verify
Verify your own completed code changes using the repo's existing infrastructure and an independent evaluator context. Use after implementing a change when you need to run unit or integration tests, check build or lint gates, prove the real surface works with evidence, and challenge the changed code for clarity, deduplication, and maintainability. If the repo is not verifiable yet, hand off to `agent-readiness`; if you are reviewing someone else's code, use `review`.
97
100%
Does it follow best practices?
Impact
89%
1.02xAverage score across 3 eval scenarios
Passed
No known issues
Quality
Discovery
100%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This is an excellent skill description that clearly articulates specific capabilities, provides explicit trigger conditions with a 'Use after...' clause, and proactively disambiguates from related skills. The inclusion of handoff guidance to 'agent-readiness' and 'review' is particularly strong for reducing conflict risk in a multi-skill environment.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions: run unit or integration tests, check build or lint gates, prove the real surface works with evidence, and challenge changed code for clarity, deduplication, and maintainability. | 3 / 3 |
Completeness | Clearly answers both 'what' (verify completed code changes using repo infrastructure and independent evaluator context) and 'when' (after implementing a change when you need to run tests, check build/lint, etc.). Also includes explicit negative triggers for handoff to other skills. | 3 / 3 |
Trigger Term Quality | Includes strong natural trigger terms users would say: 'unit tests', 'integration tests', 'build', 'lint', 'verify', 'code changes', 'maintainability'. Also includes contextual terms like 'deduplication' and 'clarity' that help match relevant requests. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive with clear boundaries: explicitly differentiates from 'agent-readiness' (for non-verifiable repos) and 'review' (for reviewing others' code). The focus on self-verification of one's own completed changes carves out a clear niche. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
100%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is an excellent skill that demonstrates strong craft across all dimensions. It provides a clear, well-sequenced verification workflow with concrete commands, specific code-quality checks, and appropriate progressive disclosure to reference files. The output example with exact fields and the handoff conditions make this immediately actionable.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is lean and efficient throughout. It assumes Claude's competence—no explanations of what tests are, what curl does, or how CI works. Every bullet earns its place with specific, actionable guidance. The principles section is tight and each point carries unique information. | 3 / 3 |
Actionability | Provides concrete commands (make verify, pnpm test, curl http://127.0.0.1:3000/health, node dist/cli.js --help), specific code-shape checks (any, unsafe as, non-null assertions), clear handoff conditions, and a complete output example with exact fields. The guidance is specific enough to execute without ambiguity. | 3 / 3 |
Workflow Clarity | The 5-step workflow is clearly sequenced with explicit ordering rationale (deterministic guardrails first, then real surface, then code shape, then adjacent risk, then verdict). Validation is built into the structure—step 1 runs guardrails, step 2 exercises real surfaces, step 4 probes failure paths. The handoff to agent-readiness when infrastructure is too weak serves as a feedback loop for blocked states. | 3 / 3 |
Progressive Disclosure | The skill provides a clear overview with well-signaled one-level-deep references to three supporting files (verification.md, evidence-rules.md, simplification.md). References are contextually placed where they're needed (evidence-rules in step 2, simplification in step 3) rather than dumped at the end. The handoffs section cleanly directs to related skills. | 3 / 3 |
Total | 12 / 12 Passed |
Validation
100%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 11 / 11 Passed
Validation for skill structure
No warnings or errors.
Reviewed
Table of Contents