Agent-Skills.md

Agent Skills: Test Executor Skill

Executes tests, analyzes test results, checks test coverage, and provides comprehensive testing status overview. Primarily for Go projects. Activates after implementing/modifying code to verify correctness, or when explicitly requested to assess test suite health.

UncategorizedID: fubira/claude-code-settings/test-executor
1

Install this agent skill to your local

pnpm dlx add-skill https://github.com/fubira/claude-code-settings/tree/HEAD/skills/test-executor

Skill Files

Browse the full folder contents for test-executor.

Download Skill

Loading file tree…

skills/test-executor/SKILL.md

Skill Metadata

Name
test-executor
Description
Executes tests, analyzes results, and reports coverage for Go and Node/Bun projects. Activates after code implementation/modification to verify correctness, or when explicitly requested.

Test Executor Skill

Execute tests, analyze results, report coverage. Supports Go and Node/Bun.

Activation Triggers

  • After code implementation/modification (automatic)
  • "test coverage", explicit test requests (manual)
  • Before commit or PR creation

Workflow

Phase 1: Detect project type and runner

| Indicator | Project type | Test command | |-----------|-------------|-------------| | go.mod | Go | go test ./... -coverprofile=coverage.out -covermode=atomic | | package.json + "test": "bun test" | Bun | bun test --coverage | | package.json + vitest/jest config | Node (vitest/jest) | npm test / npx vitest run --coverage | | Cargo.toml | Rust | cargo test | | Custom Makefile | Any | Prefer Makefile target over default |

Check package.json scripts first — respect custom commands. Coverage threshold defaults to 80% (override via project CLAUDE.md).

Phase 2: Execute

Run detected command. For coverage report:

  • Go: go tool cover -func=coverage.out
  • Bun/Vitest: coverage summary in stdout

Benchmarks only on explicit request (e.g. go test -bench=. -benchmem ./...).

Phase 3: Report

## Test Results

- Tests: X passed / Y failed
- Coverage: Z% (threshold: N%)
- Verdict: [passing / issues / failing]

### Package coverage
[per-package breakdown]

### Issues
[failures with root cause]

### Recommended actions
[specific next steps]

Failure Analysis

  • Identify root cause, not surface-level message
  • Suggest specific fixes
  • Recommend adding regression tests for the failure

Edge Cases

  • No tests found → suggest creating test files
  • Coverage generation failure → troubleshoot config
  • Flaky test → re-run to confirm
  • Build error → distinguish compile failure from test failure