A Claude Code plugin that creates a test coverage feedback loop using the undercover gem.
After running tests, Claude automatically runs undercover to find untested code changes, then writes tests for any gaps, looping until coverage passes.
Unlike total coverage percentages, undercover checks only the lines you changed, so the feedback is always relevant to the work at hand.
From marketplace:
/plugin marketplace add grodowski/undercover-claude
/plugin install undercover@undercover-ci
From local clone (development):
/plugin marketplace add /path/to/undercover-claude
/plugin install undercover@undercover-ci
Your Ruby project needs SimpleCov configured to generate coverage/coverage.json. You don't have to wire this up by hand — let the skill do it.
Inside Claude Code, run the setup prompt:
/coverage set up coverage for this project
Claude adds simplecov and undercover to your Gemfile, wires SimpleCov::Formatter::Undercover into spec_helper.rb/test_helper.rb, creates .undercover with a compare ref, runs bundle install, and runs your tests once to generate coverage data — then you're ready for the feedback loop.
The plugin installs a skill that instructs Claude to run undercover after every test run and act on the results:
undercover or undercover --include-files "app/models/foo.rb" when testing a subset of filesInvoke /coverage inside Claude Code. Type what you want to do after
the command and the skill routes to the right guide (it defaults to the feedback
loop if you say nothing):
/coverage set up coverage — SimpleCov setup instructions/coverage set up CI — CI with UndercoverCI or GitHub Actions/coverage parallel tests — coverage merging for parallel runners/coverage check this PR — read and address UndercoverCI feedback on a PRBy default undercover checks uncommitted changes only. For feature branches and CI, set a compare ref in .undercover or use the --compare flag:
-c origin/main
When coverage is reported on a GitHub PR by UndercoverCI,
the warnings live in the coverage check's annotations — which gh pr view and the
GitHub MCP server don't expose. The skill instructs Claude to pull them directly with
gh api, then fix the gaps and verify locally with --compare.
undercover-claude/
├── .claude-plugin/
│ ├── plugin.json # Plugin manifest
│ └── marketplace.json # Marketplace catalog (self-hosting)
├── skills/
│ └── coverage/
│ ├── SKILL.md # Feedback loop + router (loads on every invocation)
│ └── references/ # Guides loaded on demand
│ ├── setup.md # SimpleCov + Gemfile + .undercover + permissions
│ ├── ci.md # GitHub Actions / CircleCI / UndercoverCI / merging
│ ├── pr-feedback.md # Reading UndercoverCI PR annotations via gh api
│ └── skipping.md # :nocov:, add_filter, --exclude-files
└── README.md
MIT
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
npx claudepluginhub grodowski/undercover-claude --plugin undercoverComprehensive skill pack with 66 specialized skills for full-stack developers: 12 language experts (Python, TypeScript, Go, Rust, C++, Swift, Kotlin, C#, PHP, Java, SQL, JavaScript), 10 backend frameworks, 6 frontend/mobile, plus infrastructure, DevOps, security, and testing. Features progressive disclosure architecture for 50% faster loading.
A growing collection of Claude-compatible academic workflow bundles. Covers scientific figures, manuscript writing and polishing, reviewer assessment, citation retrieval, data availability, paper reading, literature search, response letters, paper-to-PPTX conversion, and evidence-grounded Chinese invention patent drafting. Rules are organized as reusable skill folders with explicit workflows and quality checks.
Complete creative writing suite with 10 specialized agents covering the full writing process: research gathering, character development, story architecture, world-building, dialogue coaching, editing/review, outlining, content strategy, believability auditing, and prose style/voice analysis. Includes genre-specific guides, templates, and quality checklists.
Tools to maintain and improve CLAUDE.md files - audit quality, capture session learnings, and keep project memory current.
Develop, test, build, and deploy Godot 4.x games with Claude Code. Includes GdUnit4 testing, web/desktop exports, CI/CD pipelines, and deployment to Vercel/GitHub Pages/itch.io.
Intelligent draw.io diagramming plugin with AI-powered diagram generation, multi-platform embedding (GitHub, Confluence, Azure DevOps, Notion, Teams, Harness), conditional formatting, live data binding, and MCP server integration for programmatic diagram creation and management.