Verification

Enforce evidence-based verification before claiming any task is complete

Install

skills CLI (Claude Code, Cursor, Codex and more)
npx skills add 0xranx/agentbrief --skill verification
Manual install
git clone --depth 1 https://github.com/0xranx/agentbrief.git /tmp/agentbrief
mkdir -p ~/.claude/skills
cp -r /tmp/agentbrief/briefs/code-reviewer/skills/verification ~/.claude/skills/verification

Personal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.

From the SKILL.md

> Methodology from [obra/superpowers](https://github.com/obra/superpowers) (MIT)

# Verification

Iron law: **no claims without fresh evidence.**

## The Verification Gate

Before you say "done", "works", "fixed", or "verified", you MUST:

1. **Run** the relevant command (test, build, lint, curl, etc.).
2. **Read** the full output -- not just the exit code.
3. **Confirm** the output matches the expected result.
4. **Only then** claim completion.

If you cannot run a verification command, say so explicitly. Never assume.

## What Counts as Verification

| Claim | Minimum Evidence |
|-------|-----------------|
| "Tests pass" | Paste or reference the test runner output showing green. |
| "Build succeeds" | Show the build command output with zero errors. |
| "Bug is fixed" | Show the reproduction case now producing correct output. |
| "File updated" | Read the file back and confirm the expected content. |
| "Service is running" | Hit the health endpoint and show the response. |

## Workflow

1. Finish your change.
2. Decide which claims you are about to make.
3. For each claim, run the matching verification step.
4. If any step fails, fix and re-verify. Do NOT skip ahead.
5. Report results with evidence (command + output).

## Anti-patterns to Avoid

- Saying "should work" without running anything.
- Running a command but not reading its output.
- Verifying one thing and claiming another.
Read the full skill on GitHub

Details

Skill name
verification
Path
briefs/code-reviewer/skills/verification/SKILL.md
License
MIT
Repo updated
28 March 2026

FAQ

What does the Verification skill do?

Enforce evidence-based verification before claiming any task is complete

How do I install the Verification skill?

skills CLI (Claude Code, Cursor, Codex and more): npx skills add 0xranx/agentbrief --skill verification. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.

Is Verification popular?

It lives in 0xranx/agentbrief, which has 43 GitHub stars and 0 forks.

Related skills

Typescript Advanced Types

43

0xranx/agentbrief

Master TypeScript's advanced type system including generics, conditional types, mapped types, template literals, and utility types for building type-safe applications. Use when implementing complex type logic, creating reusable type utilities, or ensuring compile-time type safety in TypeScript projects.

Software Development

Analytics Setup

43

0xranx/agentbrief

When the user needs to set up product analytics, event tracking, or says 'set up analytics,' 'add tracking,' 'GA4,' 'Mixpanel,' 'PostHog,' 'Amplitude,' 'what events should we track,' 'conversion tracking,' 'funnel tracking,' or is launching a product and needs to instrument it for data collection.

Data & Analytics

API Documentation

43

0xranx/agentbrief

Create comprehensive API documentation for developers. Use when documenting REST APIs, GraphQL schemas, or SDK methods. Handles OpenAPI/Swagger, interactive docs, examples, and API reference guides.

Software Development

Worker Integration

48

frankxai/claude-skills-library

Worker–agent integration for intelligent task dispatch and performance tracking. Use when wiring agents to a worker pool, routing tasks to workers, or instrumenting worker performance.

Productivity & Notes

Twitter Workflow

43

0xranx/agentbrief

When the user wants to draft tweets, plan threads, post to Twitter/X, or manage their Twitter presence

Productivity & Notes

Analyze Kernel Bottleneck

45

pjt222/agent-almanac

Systematically identify whether a GPU kernel is compute-bound, memory-bound, or latency-bound using roofline analysis, occupancy calculations, compute/load ratio per tile, and SASS instruction inspection. Produces a decision matrix for optimization strategy selection (cp.async, warp interleaving, tiling, double-buffering, or CuAssembler hand-tuning).

Productivity & NotesIn 1 list

Browse the most popular Claude skills, all productivity & notes skills, or write your own with the skills generator.