skills.rest
CagesThrottleUs
CagesThrottleUsCOMMUNITY
Explore all skills by this author →
GitHub

verification-before-completion

Enforces running verification commands and reading output before claiming work is complete.

Updated May 21, 2026
Search Tags:#code quality#testing#linting#verification#pre-commit#mutation testing#ci-cd
One-click install
npx skills add https://github.com/CagesThrottleUs/private-ai-harness --skill verification-before-completion-cagesthrottleus
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: verification-before-completion
Source: https://github.com/CagesThrottleUs/private-ai-harness/tree/main/skills/verification-before-completion
Command: npx skills add https://github.com/CagesThrottleUs/private-ai-harness --skill verification-before-completion-cagesthrottleus

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? AI agents and developers often claim work is done, tests pass, or bugs are fixed without actually running the verification commands, leading to broken commits and false status reports. This Skill establishes a hard gate: no completion claims without fresh verification evidence. ## Core Features & Use Cases - Evidence-First Gate: Requires identifying, running, and reading the full output of the proving command before any success claim, commit, or PR. - Language Linter Gate: Detects the project language from manifest files (pyproject.toml, package.json, go.mod, Cargo.toml, pom.xml, Gemfile) and runs the correct formatter, linter, and type checker in check mode before every commit. - Test-Strength and Quality Gates: Adds mutation testing thresholds (Stryker, PIT, mutmut) and ISO/IEC 5055 CWE-based structural quality checks on top of line coverage. - Use Case: Before committing a bug fix, the agent runs the test suite, confirms zero failures, runs ruff and mypy cleanly, verifies the regression test fails without the fix, and only then reports completion with evidence. ## Quick Start Before you commit or claim this task is done, apply the verification-before-completion gate and show me the actual command output proving it passes.

Frequently Asked Questions about verification-before-completion

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

FAQPage Schema
How do I verify code changes before committing?▼

Identify the command that proves your claim, run it fresh and completely, read the full output including exit codes, and only then state the result with evidence. For commits, also run the language-specific formatter, linter, and type checker in check mode with zero issues required.

What linter should I use for Python, TypeScript, Go, or Rust?▼

Python uses ruff for format and lint plus mypy for types. TypeScript uses Biome or ESLint/Prettier plus tsc --noEmit. Go uses gofmt and golangci-lint. Rust uses cargo fmt --check and clippy with warnings denied.

Why is line coverage not enough to prove tests are good?▼

Line coverage is gameable because a test that runs code without assertions still scores full coverage. Mutation score measures whether tests actually catch injected faults, so gates like Stryker at 50% minimum or PIT at 60% detect tautological test suites.

Can I trust a sub-agent's report that its task succeeded?▼

No. Agent success reports must be independently verified by checking the VCS diff and confirming the actual changes exist. Trusting an agent's self-report without evidence is listed as a common verification failure.

When should I not add linter suppression comments?▼

Never add noqa, eslint-disable, or clippy allow attributes just to silence output. Suppressions are only acceptable for confirmed false positives with a documented reason; otherwise fix the violation and re-run until clean.

Related Skills

Explore More Software Engineering Skills→
oabdelmaksoudoabdelmaksoud
4

verification-before-completion

Require fresh verification commands and output evaluation before completion claims.

Community
Intermediate
roycolumbia-coderoycolumbia-code

gsd-ship

Automates GitHub pull request creation, review triggering, and merge tracking.

Community
Intermediate
jjeremycaijjeremycai

review

Dispatch three parallel code-reviewer subagents to inspect diffs, commits, and pull requests.

Community
Basic
metalspawnmetalspawn

execute

Execute code tasks through worker implementation, parallel reviews, and commit generation.

Community
Advanced