Test Coverage Mcp
Connect your agent to LCOV output so it can see uncovered lines and suggest targeted tests without you pasting full coverage dumps into every chat.
Overview
io.github.goldbergyoni/test-coverage-mcp is an MCP server for the Ship phase that analyzes LCOV test-coverage files so coding agents can work coverage-aware without loading entire reports into context.
What is this MCP server?
- Parses LCOV coverage files and exposes coverage context to MCP-connected coding agents
- Designed to make agents coverage-aware while avoiding token-heavy raw LCOV pastes
- stdio npm package test-coverage-mcp (registry v1.1.0) for local MCP clients
- Fits CI workflows where jest, vitest, nyc, or similar tools already emit LCOV
- Helps agents prioritize uncovered branches and files during test-gap conversations
- Server schema version 1.1.0 per MCP registry metadata
- Single npm package identifier test-coverage-mcp with stdio transport
- Source repository: github.com/goldbergyoni/test-coverage-mcp
Community signal: 41 GitHub stars.
What problem does it solve?
Pasting LCOV or coverage HTML into every agent session burns tokens and still leaves the model guessing which lines and branches are actually uncovered.
Who is it for?
Indie developers using MCP-enabled agents who already emit LCOV from CI or local test runs and want coverage-informed test suggestions during Ship.
Skip if: Teams without LCOV-based workflows, non-developer projects, or anyone expecting the server to run tests or enforce coverage thresholds by itself.
What do I get? / Deliverables
After you register the server and point it at your LCOV output, your agent can discuss gaps, priorities, and test ideas using real coverage signals instead of improvised assumptions.
- MCP-accessible interpretation of LCOV coverage for agent-driven conversations
- More targeted test suggestions aligned to uncovered files, lines, or branches
- Lower token use versus repeatedly inlining full coverage reports in prompts
Recommended MCP Servers
Journey fit
Test coverage analysis belongs in Ship because it supports safe release decisions after you have a test suite and CI-generated LCOV artifacts. Testing is the canonical shelf: the server’s purpose is interpreting LCOV coverage data to guide what to test next, not writing features or shipping marketing.
How it compares
MCP integration for LCOV coverage insight, not an agent skill checklist and not a test framework or coverage gate.
Common Questions / FAQ
Who is test-coverage-mcp for?
It is for solo builders and small teams who use AI coding agents with MCP and already generate LCOV files from their test toolchain.
When should I use test-coverage-mcp?
Use it during Ship when you are reviewing coverage before merge, planning tests for untested code, or refactoring modules where you need the agent to see actual LCOV-backed gaps.
How do I add test-coverage-mcp to my agent?
Install the npm package test-coverage-mcp, configure it as a stdio MCP server in your client (for example Claude Code or Cursor), and ensure your test runs produce LCOV the server can read.