lee-fuhr Part of Claude Code operator skills: a collection of skills for running a real Claude Code setup.
Audit QA and testing with expert personas
Run a full serial QA audit using 22 deep expert persona frameworks, fixing critical issues and verifying each framework before proceeding to the next.
22 testing frameworks, each loaded as a deep expert persona into its own agent context. Serial execution: one framework → fix critical issues → verify → next. Each agent thinks like a 20-year QA specialist in that specific framework.
Domain: Testing. Expert lens: Is it tested? Would a QA lead approve?
Modes
This skill responds to three modes based on the args passed. Interpret intent loosely: fuzzy matching, not exact phrasing.
Mode 1: Full serial audit (no args or "run all")
Triggers: /qq-audit-testing, "run all", "full audit", "everything", no args at all.
Mode 2: Single framework (framework name + optional scope)
Triggers: /qq-audit-testing mutation testing, "just run boundary value on the parser", "test isolation on the API layer", etc.
Match the framework name fuzzily: "pyramid", "test pyramid", "pyramid balance" should all match Test Pyramid Balance. "flaky", "flaky tests", "quarantine" should all match Flaky Test Detection. If ambiguous, show the 2-3 closest matches and ask.
Mode 3: List frameworks (help/list/discovery)
Triggers: "list", "what do you have", "what frameworks", "help", "which ones", "show me", etc.
Show the framework table from §Framework inventory below, then ask which one(s) to run.
Smart interview (runs before any audit)
DO NOT ask dumb questions. Before asking anything, gather what you already know:
- Check conversation context: What product are we working on? What files have been discussed? What has the user been complaining about?
- Check project CLAUDE.md: Product description, tech stack, target audience.
- Check recent session state: What was just built or changed?
Pre-fill and present assumptions:
"Here's what I know going in:
- Product: [name] ([description from context])
- Tech stack: [languages, frameworks, test runners, inferred from project]
- Current test state: [test count, coverage %, CI status, from recent runs]
- Known pain points: [what the user has mentioned or what recent changes suggest]
- Scope: [full codebase / specific module / specific test suite]
Anything wrong or missing?"
Use AskUserQuestion with multiple choice ONLY for genuine gaps. For example, if you truly can't tell what test runner is in use, ask. If you can infer it, state the inference.
Interview output becomes the audit context: passed to every framework agent so they audit with purpose, not generically.
Execution flow (full audit)
Phase 1: Context gathering
- Run smart interview (above)
- Read the product's test files, config, and CI pipeline to understand scope
- Run existing test suite to establish baseline
Phase 2: Serial framework execution
For each framework (in order 1-22):
Spawn one agent with these instructions:
- "Read the framework subskill file at
~/.agents/skills/qq-audit-testing/frameworks/[NN-name].md" - "You ARE the expert described in that document. Think like them, not like a generalist."
- "Audit [product name] focusing on [scope]. Context: [interview output]"
- "Previous frameworks found these issues: [cumulative findings list]. Do NOT re-report duplicates."
- "Score 1-10, list findings with file:line references, fix critical issues (score < 7), write report."
- "Read the framework subskill file at
Collect the agent's output:
- Score for this framework
- Findings (new, not duplicates)
- Fixes applied
- Items flagged for downstream frameworks
Verify build if fixes were applied (run test suite or equivalent)
Update cumulative state:
- Add findings to the dedup list
- Update the running score card
- Note cross-framework flags for upcoming frameworks
Move to next framework
Phase 3: Synthesis
- Run framework #22 (Data Integrity/Migration Testing) as the final evaluation
- Compile consolidated report:
- Overall testing health score (average of all frameworks)
- Per-framework scores
- All findings grouped by severity
- Fixes applied (with file:line)
- Remaining items for future rounds
- Compare to previous audit rounds if they exist
Phase 4: Report
Save to project's data directory:
data/audit-testing-[date].md: full reportdata/audit-testing-[date]-summary.md: scores + critical findings only
Framework inventory
| # | Framework | Expert lens | Subskill file |
|---|---|---|---|
| 1 | Test Pyramid Balance | Healthy ratio of unit:integration:e2e. Fast feedback at the base, confidence at the top | 01-test-pyramid.md |
| 2 | Meaningful Code Coverage | Assertion density, not just line execution: does your coverage number actually mean anything? | 02-code-coverage.md |
| 3 | Boundary Value Analysis | Off-by-one, empty, null, max, zero, negative: bugs cluster at the edges of valid input ranges | 03-boundary-value.md |
| 4 | Equivalence Partitioning | Test cases represent input classes: one value from each partition proves the whole partition works | 04-equivalence-partitioning.md |
| 5 | Happy Path / Sad Path Coverage | Every failure mode tested: optimistic-only testing is how bugs reach production | 05-happy-sad-paths.md |
| 6 | Test Isolation / Independence | No shared state, any-order execution: tests that depend on each other are tests that lie to you | 06-test-isolation.md |
| 7 | Mutation Testing | Tests that actually detect bugs: mutate the code, see if the tests catch it | 07-mutation-testing.md |
| 8 | Contract Testing | Consumer-driven API contracts verified on both sides: no more "it works on my machine" between services | 08-contract-testing.md |
| 9 | Visual Regression Testing | Screenshot comparison catches CSS regressions: the bugs that functional tests structurally cannot see | 09-visual-regression.md |
| 10 | Accessibility Testing Automation | WCAG violations caught in CI: automated axe-core/pa11y scans as the first line of accessibility defense | 10-a11y-testing-automation.md |
| 11 | Load/Stress Testing | System performs under expected and peak load: find the breaking point before your users do | 11-load-stress-testing.md |
| 12 | Chaos/Resilience Testing | Handles infrastructure failures gracefully: break things intentionally so they don't break you in production | 12-chaos-resilience.md |
| 13 | Smoke/Sanity Test Suite | Fast critical path verification after deployment: the 2-minute check that saves 2-hour incidents | 13-smoke-sanity.md |
| 14 | Test Data Management | Factories over fixtures, no shared mutable state: test data strategy determines test reliability | 14-test-data-management.md |
| 15 | Flaky Test Detection and Quarantine | Identified, quarantined, tracked: flaky tests erode trust and must be managed as defects, not nuisances | 15-flaky-test-detection.md |
| 16 | Cross-Browser/Cross-Device Testing | Works across target browsers and devices: not just the browser on the developer's laptop | 16-cross-browser-device.md |
| 17 | Error Scenario Simulation | Fault injection at the application level: API failures, timeouts, auth expiration, network errors, all tested deliberately | 17-error-scenario-simulation.md |
| 18 | Regression Test Effectiveness | Historical defect escape rate measured: does the test suite actually catch regressions, or does it just feel like it does? | 18-regression-effectiveness.md |
| 19 | Test Readability/Maintainability | Arrange-Act-Assert, descriptive names, self-documenting tests: tests are documentation that happens to be executable | 19-test-readability.md |
| 20 | CI Pipeline Speed and Reliability | Under 10 minutes to green, failures are real: the CI pipeline is a feedback loop, and slow loops kill quality | 20-ci-pipeline-speed.md |
| 21 | Feature Flag Testing | Both on/off states tested, conflicting combinations covered: feature flags multiply your test surface, not simplify it | 21-feature-flag-testing.md |
| 22 | Data Integrity/Migration Testing | Schema migrations tested forward and backward: because an untested migration is a production outage waiting for Friday at 5 PM | 22-data-migration-testing.md |
Key principles
- Serial, not parallel: 70% of findings duplicate across frameworks. Serial means each round finds genuinely new issues after fixes.
- Fix before moving on: don't accumulate a findings list. Fix each framework's criticals before the next audit.
- Expert persona, not checklist: each agent IS the specialist. They reason from principles, not rules.
- Hold every fix to a real quality bar: no retrospective-only tests, red/green/refactor discipline, and unhappy paths tested before happy paths.
- Code-level evidence: every finding must reference specific files and lines. No hand-waving.
- Multi-round: after all 22 frameworks, run the full cycle again. Scores increase each round until plateau.
- Dedup across frameworks: each agent receives cumulative findings so they don't re-report known issues.