supercov — Claude Code skill: what it does, how to install
A Claude Code skill that inspects test coverage and code structure using the supercov CLI. It pinpoints missing test branches and writes targeted checks across eight languages.
MIT ★ 167 updated 2026-11-13
npx skills add supercorp-ai/supercov --skill supercov -a claude-codeWhat supercov does
The skill wraps the existing project test suite inside npx supercov -- <test command> without editing setup files. After evaluating test coverage, the agent queries unverified files via npx supercov runs latest gaps and reads target paths. It creates a single targeted test, reruns the verification, and compares coverage numbers before and after changes.
Static issues and design defects are tracked through npx supercov quality, while npx supercov quality patch checks incoming edits.
Running via npx requires Node.js 22 or newer, otherwise the standalone supercov executable must be present in the system. The quality checking command transmits files to TypeSafe and fails without a valid TYPESAFE_API_KEY token.
What it helps with
- detect uncovered lines in the test suite
- write focused unit tests for untested files
- track code smell regressions in patches
- compare test coverage metrics before and after edits
Example prompts for the agent
- Measure current test coverage with
npx supercov -- npm test - List coverage gaps with
npx supercov runs latest gaps --limit 5and cover the first one - Run
npx supercov quality patchto verify recent code changes
How to install supercov
1Skills CLI (Vercel)
npx skills add supercorp-ai/supercov --skill supercov -a claude-code2Manually: put the files in place
~/.claude/skills/supercov/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/supercov/
1Skills CLI (Vercel)
npx skills add supercorp-ai/supercov --skill supercov -a codex2Manually: put the files in place
.agents/skills/supercov/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add supercorp-ai/supercov --skill supercov -a cursor2Manually: put the files in place
.cursor/skills/supercov/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install supercorp-ai/supercov plugins/supercov/skills/supercov2Manually: put the files in place
.github/skills/supercov/SKILL.md1Skills CLI (Vercel)
npx skills add supercorp-ai/supercov --skill supercov -a gemini-cli2Platform CLI
gemini skills install https://github.com/supercorp-ai/supercov --path plugins/supercov/skills/supercovManually: put the files in place
.devin/skills/supercov/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add supercorp-ai/supercov --skill supercov -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about supercov
How do I install supercov in Claude Code?
Run npx skills add supercorp-ai/supercov --skill supercov -a claude-code. Alternatively, place the skill definition file at ~/.claude/skills/supercov/SKILL.md.
What does supercov do?
It runs project test suites, evaluates coverage deficits, and creates targeted test cases without altering application source code.
What is required to run supercov?
It requires Node.js 22 or newer, or a standalone supercov binary. Analyzing code quality also requires setting the TYPESAFE_API_KEY environment variable.
Source text
SKILL.md · MIT
--- name: supercov description: Measures test coverage and code quality in a repository with the supercov CLI, and turns what it finds into small, focused tests or fixes. Use when the user asks to add or improve tests, find untested code or raise coverage, or find what to refactor. Works with JavaScript, TypeScript, Python, Ruby, Rust, Go, Java and Kotlin projects. license: MIT --- # Supercov Supercov runs the project's own test command and changes no source, tests or configuration. `npx supercov docs <topic>` prints the guide for the installed version; read it for anything not shown here. Topics: `agent-loop`, `quality`, `supported-suites`, `troubleshooting`. ## Add a test 1. Measure the full suite: `npx supercov -- <test command>`, for example `npx supercov -- npm test`. 2. List targets with `npx supercov runs latest gaps --limit 5`, pick one, and inspect it with `npx supercov runs latest file <path>`. 3. Write one focused test. Change only tests unless the user asks otherwise. 4. Rerun the same command and report the test and the coverage before and after. ## Quality - `npx supercov quality` ranks files by code smells, and `npx supercov quality patch` reports what a change introduced. - It sends source files to TypeSafe using the `TYPESAFE_API_KEY` environment variable. A user who set the key has opted in, so run it without asking again. If the key is missing the command says so: tell the user how to set it, then answer from reading the code and say Supercov did not check it. Without Node.js 22 or newer, install Supercov as the language guide at https://supercov.com/docs says, and run `supercov` instead of `npx supercov`.
© supercorp-ai · MIT · supercorp-ai/supercov