# Supercov

> Coverage for coding agents and software factories. Supports JavaScript, TypeScript, Rust, Python, and Ruby. Supercov measures an existing test command; the coding agent writes tests and builds the assertion map.

## Start here for agents

Prefer `supercov docs agent-loop`, `supercov docs assertion-agent` and `supercov docs cli` from the installed version. Use the native invocation for the project: `npx supercov` for JS/TS, `uvx --from supercov-cli supercov` for Python, or installed `supercov` for Ruby and Rust.

- [Agent instructions](https://supercov.com/docs/agents): entry point and available references.

JavaScript example (replace npm test with the project's actual full test command):
Run: `npx supercov -- npm test`
Query: `npx supercov runs latest gaps --limit 10`
Compare: `npx supercov diff <previous-run-id> latest`
Queries support `--json`; collection queries are paginated.

## Released Markdown reference

Mirrored from supercorp-ai/supercov at commit 5791b7587adfea9bc10662011f41a24143f3aec6. These are the full technical guides, not the shorter human-facing website guides. If the installed version differs, read its bundled documentation instead.

- [Getting started](https://supercov.com/reference/getting-started.md)
- [Agent workflow](https://supercov.com/reference/agent-loop.md)
- [Troubleshooting](https://supercov.com/reference/troubleshooting.md)
- [CLI reference](https://supercov.com/reference/cli.md)
- [Supported languages and test suites](https://supercov.com/reference/supported-suites.md)
- [Understanding coverage](https://supercov.com/reference/coverage-model.md)
- [Assertion coverage](https://supercov.com/reference/assertions.md)
- [Mapping assertions with an agent](https://supercov.com/reference/assertion-agent.md)
- [Investigating assertion evidence](https://supercov.com/reference/assertion-evidence.md)
- [Assertion map format](https://supercov.com/reference/assertion-maps.md)
- [Runs and evidence](https://supercov.com/reference/evidence.md)
- [Files, privacy, and cleanup](https://supercov.com/reference/workspace-isolation.md)
- [Trusting a result](https://supercov.com/reference/verification.md)
- [Speed and storage](https://supercov.com/reference/performance.md)

## Human guides

- [Getting started](https://supercov.com/docs/getting-started): Ask your agent to add one test, then review what changed.
- [Installation](https://supercov.com/docs/installation): Install Supercov for your language and run your first measurement.
- [Supported languages](https://supercov.com/docs/supported-languages): Use Supercov with your project's language and test suite.
- [Working with your agent](https://supercov.com/docs/agent-loop): Choose what to test, how long to spend and when to stop.
- [JavaScript and TypeScript test coverage](https://supercov.com/docs/javascript): Set up Supercov with your agent, or run it yourself with Vitest, Jest, node:test or Playwright.
- [Python test coverage](https://supercov.com/docs/python): Set up Supercov with your agent, or run it yourself with pytest or unittest.
- [Ruby test coverage](https://supercov.com/docs/ruby): Set up Supercov with your agent, or run it yourself with RSpec, Minitest or Rails tests.
- [Rust test coverage](https://supercov.com/docs/rust): Set up Supercov with your agent, or run it yourself with Cargo tests.
- [Lines and branches](https://supercov.com/docs/coverage-model): See which parts of your code ran during the tests.
- [MC/DC](https://supercov.com/docs/mcdc): Check whether tests exercise each part of a decision independently.
- [Reading the results](https://supercov.com/docs/verification): Understand the coverage report and review your agent's changes.
- [Assertion coverage](https://supercov.com/docs/assertions): See what the tests check, not just what they run.
- [Troubleshooting](https://supercov.com/docs/troubleshooting): Find out why a run failed or a result needs another look.
- [Files and privacy](https://supercov.com/docs/workspace-isolation): Know what stays local and what can change in your project.

## Web technical reference

- [Supported languages and test suites](https://supercov.com/docs/supported-suites): Check the detail available from each runner and its measurement limits.
- [Mapping assertions with an agent](https://supercov.com/docs/assertion-agent): Build and update the links between assertions and the source they check.
- [Investigating assertion evidence](https://supercov.com/docs/assertion-evidence): Investigate missing credit, stale mappings and execution details.
- [Assertion map format](https://supercov.com/docs/assertion-maps): Look up map fields, source references and validation rules.
- [CLI reference](https://supercov.com/docs/cli): Find commands for measuring, querying, comparing and checking runs.
- [Agent instructions](https://supercov.com/docs/agents): Give your agent the instructions bundled with its installed version.
- [Runs and evidence](https://supercov.com/docs/evidence): Revisit, filter, compare and retain local runs.
- [Speed and storage](https://supercov.com/docs/performance): Understand run times, repeated measurements and local disk usage.

## Blog

- [Assertion coverage without mutation testing](https://supercov.com/blog/assertion-coverage-without-mutation-testing): 100% code coverage can miss a wrong result. See how assertion coverage finds missing checks in passing tests, and how it differs from mutation testing.
- [Claude Code testing: unit tests, browser tests, and MCP](https://supercov.com/blog/claude-code-testing): A practical guide to Claude Code testing: use your existing suite, write unit and integration tests, connect Playwright MCP, and check what the tests miss.
- [MC/DC coverage for JavaScript](https://supercov.com/blog/mcdc-coverage-javascript): See how MC/DC finds an untested condition in JavaScript, even with 100% line and branch coverage. A small example with real Supercov output.

## Comparisons

- [Vitest coverage: V8 vs Istanbul](https://supercov.com/compare/vitest-coverage): Compare Vitest's V8 and Istanbul coverage providers, set up source coverage, and see when MC/DC and assertion coverage add useful evidence.
- [c8 vs nyc: JavaScript coverage compared](https://supercov.com/compare/c8-vs-nyc): Compare c8 and nyc coverage with the same JavaScript fixture. See how collection, source maps, and coverage criteria affect the result.
- [Vitest vs Jest: which test runner should you choose?](https://supercov.com/compare/vitest-vs-jest): Compare Vitest and Jest for Vite, TypeScript, mocks, browser testing, and coverage. Know what to check before migrating an existing test suite.
- [Ruby coverage: SimpleCov vs Supercov](https://supercov.com/compare/simplecov): Compare SimpleCov and Supercov for Ruby line, branch, per-test, MC/DC, and assertion coverage, with a concrete missing-test example.
- [Python coverage: coverage.py, SlipCover, and Supercov](https://supercov.com/compare/python-code-coverage): Compare Python coverage tools: coverage.py with pytest-cov, SlipCover, and Supercov. See per-test contexts, MC/DC, and assertion coverage side by side.
- [Rust coverage: cargo-llvm-cov, Tarpaulin, and Supercov](https://supercov.com/compare/rust-code-coverage): Compare Rust coverage tools, including cargo-llvm-cov's nightly MC/DC support, Tarpaulin, and Supercov's per-test and assertion coverage.

## Source

- [GitHub](https://github.com/supercorp-ai/supercov)
- [npm](https://www.npmjs.com/package/supercov)