Find the perfect skill

Testing

Run a brepjs-cad skill eval in Claude Code: re-author playground examples, have a blind judge grade them, and propose SKILL.md fixes. Optional verify mode uses a precision/recall harness, no API key.

Testing

Analyze code coverage gaps and generate targeted tests for uncovered branches, functions, and edge cases.

Testing

Audit tool integrations for deterministic behavior, error contracts, and logging before agents depend on them. Use when adding or updating tools or MCP servers.

Testing

Validates that an FTC skill plugin follows marketplace requirements and best practices.

Testing

Browser automation CLI for AI agents. Navigate pages, fill forms, click elements, take screenshots, extract data, and test web apps via accessibility-tree snapshots.

Testing

Dynamically discovers closed GitHub issues for fixed bugs, verifies each fix is still in place, and reports any regressions found.

Testing

Per-game audit of Fallout 4 compatibility: BA2 archives, half-float vertices, BGSM materials, and M49 precombined geometry (CSG).

Testing

Systematically QA test a web application, find bugs, and fix them iteratively. Supports quick, standard, and exhaustive tiers with health scoring.

Testing

Python quality check using pytest, mypy, ruff, black, and isort. Enforces 90%+ test coverage and mypy strict mode.

Testing

Performs a supervised dry-run of the specified skill in a disposable sandbox with fixed inputs, checks output invariants, and reports PASS/FAIL without touching real areas.

Testing

Deep audit of ByroRedux readiness for Fallout: New Vegas, hunting regressions and missing polish on the reference title.

Testing

Validates proposal deliverables: script quality, JSON conversion accuracy, and HTML visual quality. Supports SPARK-6, S-E-P, Win Theme, pattern diversity, and PPTX mode.