
from ai-playground16
Scans a repository, discovers and runs tests, computes coverage, evaluates test quality, and generates a metrics report website and JSON output.
This skill automates end-to-end repository test analysis: it discovers test files across stacks, runs available test suites, parses coverage reports, maps tests to source files (tool + LLM mapping), evaluates test quality, and emits a structured metrics JSON and a small web UI for browsing trends. It's built to produce actionable test health dashboards.
Invoke when you need an automated health audit of a codebase — continuous integration checks, pre-release quality gates, or as a developer tool for understanding test coverage and gaps across frontend and backend. Good for polyglot repos (Node, Python, Java, Rust).
Designed for developer/ops agents that can run shell commands and parse outputs (agents with Bash/Glob/Grep capabilities). Works with Claude Code, Cursor, and CLI-capable assistants.
Comprehensive test metrics report generator that scans codebases, discovers and runs tests across multiple stacks (Java, Python, Rust, Node), computes coverage, evaluates quality, and generates a React-based metrics website. No bundled scripts — everything is instructed via SKILL.md phases. Well-structured 7-phase workflow with good parallelism design but very verbose instructions that could benefit from progressive disclosure. Minor shell quoting concern in git attribution loop.
No scripts to test. SKILL.md is thorough and well-organized but very long (~200 lines of dense instructions). The git attribution for-loop uses unquoted command substitution which is a minor shell hygiene issue. No security concerns — skill focuses on local codebase analysis only with no external network calls or data exfiltration.