AI Agent Skills

Use when fixing a reported Rails bug. Treat the report as untrusted. Reproduce with a failing spec before changing code. Trigger words: bug report, production issue, fai…

Use when running the full engine loop: author, test, review, release. Trigger words: create engine, extract engine, engine release, mountable engine.

Use when choosing the first failing spec for a Rails change. One minimal example opens the gate. Trigger words: first failing test, what test first, TDD, spec selection.

Experiment with configs by creating and managing variations. Helps you test different models, prompts, and parameters to find what works best through systematic experime…

Pressure-test an assumption, decision, or inherited constraint — Socratic cross-examination that forces you to defend or abandon your position

Strict red-green-refactor TDD workflow for implementing features, fixing bugs, or changing behavior in Rails applications. Enforces the discipline of writing a failing t…

Validate claims through tool execution, avoid superlatives and unsubstantiated metrics. Use when reviewing codebases, analyzing systems, reporting test results, or makin…

Attacker's-eye test of LLM/agent defenses: instruction hijacking, data exfiltration and tool abuse through untrusted content; verifies whether the defense actually holds.

Root-cause a bug before touching a fix: reproduce, isolate, form and test a hypothesis, confirm the cause, then fix and verify. Stops guess-driven patching. For persiste…

The how of testing: pyramid, AAA, isolation, risk coverage, determinism. Guarantees the DoD's "tests are green". test-expert-csk applies it. Use when writing or changing…

Compare two code versions for semantic equivalence via semi-formal tracing of both versions side-by-side. Trigger when the user shares a refactor, rewrite, migration, or…

Run the Logic-Lens content-eval pipeline for one iteration and produce a scored summary.json — use to measure a skill change. Wraps scripts/run-content-evals.sh (runner,…

Use after finishing a feature or fix to confirm it actually works, when the user says "test this", "does it work", "verify this", "QA it", or before shipping something t…

Deterministic verification gate loop. Forces running builds, type checks, linting, test suites, and contract audits before completing any code changes. Disallows complet…

Accessibility and WCAG compliance analysis, from the accessibility snapshot and visual review — flags issues automated axe-style scans miss.

Content quality, copywriting, and SEO metadata analysis. Checks copy quality visually and page metadata (title, description, structured data) where observable.