skills.rest
nsollazzo
nsollazzoCOMMUNITY
Explore all skills by this author →
Website𝕏@n_sollazzoGitHub

verify

Runs project tests and exercises changed code to confirm fixes work with command-output evidence.

Updated Jun 15, 2026
Search Tags:#testing#test automation#bug fix#evidence#verify#ci#runtime validation
One-click install
npx skills add https://github.com/nsollazzo/ship-kit --skill verify-nsollazzo
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: verify
Source: https://github.com/nsollazzo/ship-kit/tree/main/.agents/skills/verify
Command: npx skills add https://github.com/nsollazzo/ship-kit --skill verify-nsollazzo

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Code changes often look correct on review but fail at runtime. This Skill closes that gap by actually executing the project's tests and exercising the changed behavior, producing concrete evidence instead of assumptions. ## Core Features & Use Cases - Project recipe discovery: Finds the repo's own test and run commands from CLAUDE.md, README, package.json, Makefile, CI workflows, and similar sources before running anything. - Runtime exercise: Boots the app, hits the touched endpoint, runs the CLI subcommand, or reproduces the original bug to confirm the fix in real behavior. - Evidence-based reporting: Reports the exact commands run and their output, distinguishing "tests pass" from "feature exercised and observed working". - Use Case: After fixing a login bug, ask the agent to verify the change — it runs the test suite, reproduces the original failure scenario, and confirms the bug is gone with quoted output. ## Quick Start Verify that the fix for the login timeout bug actually works by running the project's tests and reproducing the original issue.

Frequently Asked Questions about verify

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

FAQPage Schema
How do I verify a code fix actually works?▼

Run the project's own test suite, then exercise the changed behavior at runtime — start the server, hit the endpoint, or reproduce the original bug. This Skill discovers the repo's test commands automatically and reports the exact commands and output as evidence.

How to test a change without knowing the project's test framework?▼

The Skill discovers the project's recipe by checking CLAUDE.md, README, package.json, Makefile, pyproject.toml, cargo config, and CI workflow files. CI configuration typically reveals the real commands the project uses.

Does this verification work with any programming language?▼

Yes, it is language-agnostic and assumes no specific test framework. It follows whatever commands the project defines, whether npm test, pytest, cargo test, go test, or make test.

When is a runtime check skipped during verification?▼

For pure docs or config changes with no runtime surface, running tests or a lint/build is sufficient. The skip is always reported explicitly with the reason, never silently omitted.

What happens when verification fails in a pipeline?▼

The run stops and fails loud, reporting the failing command and its output verbatim. It never reports success when something errored, since surfacing real failures is the purpose of the gate.

Related Skills

Explore More Software Engineering Skills→
skenklokskenklok

pr-review

Review pull requests against CLAUDE.md, prd.json, or design docs.

Community
Advanced
andymaiandymai
77

git-pr-workflow

Resolves Git workflow blockers for the brepjs repository.

Community
Advanced
civitaicivitai
7.2k

cleanup

Runs comment-review and docs-drift-review lanes over session changes and applies the fixes.

Official
Intermediate
jakubbartnikjakubbartnik

systematic-debugging

Diagnose software bugs through a four-phase root cause investigation workflow.

Community
Intermediate