Gauntlet Hardener — Testing & QA agent for Claude Code
Stage 5 of the agent gauntlet.
How to install Gauntlet Hardener
Installs to ~/.claude/agents/francis-infotrack-francisilagan-claude-skills-gauntlet-hardener.md
mkdir -p ~/.claude/agents && curl -fsSL https://raw.githubusercontent.com/francis-infotrack/francisilagan-claude-skills/HEAD/agents/gauntlet-hardener.md -o ~/.claude/agents/francis-infotrack-francisilagan-claude-skills-gauntlet-hardener.md Restart Claude Code, or start a new session, for it to be picked up.
What Gauntlet Hardener does
name: gauntlet-hardener description: Stage 5 of the agent gauntlet. Merciless mutation testing over the story's diff — flip operators, mutate constants and conditionals, and kill every surviving mutant by adding a test that catches it. Proves the test suite actually constrains the code, not just covers it. model: opus
You are the **hardener** — the fifth stage of the agent gauntlet. Coverage says the tests *execute* the code; you prove they *constrain* it. A mutant that survives means a
Alternatives in Testing & QA
- Bowser QA Agent — UI validation agent that executes user stories against web apps and reports pass/fail results with screenshots 222 ★
- Quality Reviewer — Use AFTER spec-reviewer passes to review code quality 60 ★
- Test Isolation Reviewer — Use this agent after adding or modifying any file in tests/ to verify the test will pass in CI (no parent-proj 56 ★
Full documentation available on GitHub
View Source RepositoryRelated Agents
Mutation Test Auditor
Verification Quad Agent 4. Validates test rigor against pytest-conventions and runs worktree-isolated mutation
Sb Experimenter Low
strictlybetter experimenter pinned to LOW reasoning effort. MUST BE USED by /strictlybetter:run for the config
Test Review Gates
Runs the objective test-verification gates (red-green implementation-removal + mutation testing) for ONE shard
Mutation Analyst
Analyses test suite strength via Stryker.NET mutation testing and reports surviving mutants
Idea Assessor
Spec-Kit idea assessor. Runs one assessment stage per call (intake, research, define, shape, decide) and ends
Mutation Runner
Applies a single code mutation in an isolated worktree and runs the test to check if it catches the bug. Used
Related Skills
Kill Mutants
Analyze surviving mutants from a mutation testing run and write targeted unit tests to kill them. Re-runs muta
Mutant Killers
Act as a rigorous senior Perl SDET specializing in mutation testing. Write a comprehensive set of tests in ./t
Mutate
Run the Playbook §4 mutation pass on critical modules — the real anti-performative metric — and kill survivors