verification
A rigorous verification skill that uses discriminating tests and data tracing to validate system behavior and catch bugs.
Install
mkdir -p .claude/skills/verification && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/19309" && unzip -o skill.zip -d .claude/skills/verification && rm skill.zipInstalls to .claude/skills/verification
Activation
This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.
Use when: verifying an implementation, auditing the codebase, reviewing a finished change, doing a pre-commit scan, or when asked to 'check everything' in SAKIGo. Evidence-driven verification procedure that replaces checklist ceremony with discriminating tests. Covers: running test suites, tracing tiny examples, auditing with subagents, filtering false positives, and SAKIGo-specific invariants (perspective flips, pass logit, D4 equivariance, masks).Key capabilities
- →Run test suites for Python and Rust
- →Trace tiny examples end-to-end
- →Audit codebases using subagents
- →Verify SAKIGo-specific invariants
- →Record risks in AI/Issues.md
How it works
The skill uses evidence-gathering through test execution and manual tracing to distinguish between functional and broken code states.
Inputs & outputs
When to use verification
- →Audit codebase for specific bugs
- →Verify implementation after a code change
- →Perform a pre-commit verification scan
About verification
Replaces checklists with concrete evidence-gathering. It runs test suites, performs trace-based debugging, and audits code to distinguish between functional and broken states.
Use when: verifying an implementation, auditing the codebase, reviewing a finished change, doing a pre-commit scan, or when asked to 'check everything' in SAKIGo. Evidence-driven verification procedure that replaces checklist ceremony with discriminating tests. Covers: running test suites, tracing t
When not to use it
- →Re-reporting known watchlist items
- →Fixing design decisions during verification
- →Delegating audits to Haiku-class subagents
Prerequisites
Limitations
- →Subagent findings have a high false-positive rate
- →Clean scans with untested behavior are insufficient
How it compares
It replaces checklist-based reviews with discriminating tests that falsify assumptions about code correctness.
Compared to similar skills
verification side by side with the closest alternatives in the catalog.
| Skill | Installs | Updated | Safety | Difficulty |
|---|---|---|---|---|
| verification (this skill) | 0 | 3mo | No flags | Advanced |
| python-testing-patterns | 77 | 5mo | Review | Intermediate |
| fix-bug | 11 | 9mo | Review | Intermediate |
| test-fixing | 1 | 11mo | Review | Intermediate |
Try saying
Example prompts that trigger this skill in your AI assistant.
You might also like
python-testing-patterns
wshobson
Implement comprehensive testing strategies with pytest, fixtures, mocking, and test-driven development. Use when writing Python tests, setting up test suites, or implementing testing best practices.
fix-bug
tddworks
Guide for fixing bugs in ClaudeBar following Chicago School TDD and rich domain design. Use this skill when: (1) User reports a bug or unexpected behavior (2) Fixing a defect in existing functionality (3) User asks "fix this bug" or "this doesn't work correctly" (4) Correcting behavior that violates the user's mental model
test-fixing
mhattingpete
Run tests and systematically fix all failing tests using smart error grouping. Use when user asks to fix failing tests, mentions test failures, runs test suite and failures occur, or requests to make tests pass.
fixing-bugs-systematically
CaptainCrouton89
Diagnose and fix bugs through systematic investigation, root cause analysis, and targeted validation. Use when something is broken, errors occur, performance degrades, or unexpected behavior manifests.
moai-workflow-testing
modu-ai
Comprehensive development workflow specialist combining DDD testing, debugging, performance optimization, code review, PR review, and quality assurance into unified development workflows
investigate
MadAppGang
Unified entry point for code investigation. Auto-routes to specialized detective based on query keywords. Use when investigation type is unclear or for general exploration.