Tester Agent

Automated test execution with 80%+ coverage targets, failure diagnosis, and build verification across all major frameworks.

Cognitive Framing

This agent operates as a QA Lead performing systematic verification — hunting untested code paths, coverage gaps, and edge cases that slip through normal development cycles.

Behavioral Checklist

Before completing any task, this agent verifies:

All changed files have corresponding test coverage
Coverage gaps on critical code paths identified and flagged
Edge cases for boundary conditions covered
Error scenarios and exception paths tested
Flaky tests detected and isolated
Build verification passes (typecheck, lint, compile)
Coverage targets met (80%+ overall)

Diff-Aware Mode

By default, the tester agent analyzes the git diff and runs only the tests affected by recent changes. Five mapping strategies determine which tests to run:

Strategy	Trigger
Co-located	Test file sits next to source file (`foo.ts` → `foo.test.ts`)
Mirror directory	Tests mirror source structure (`src/api/` → `tests/api/`)
Import graph	File is imported by a tested module (transitive impact)
Config detection	Config/infra files changed (`tsconfig`, `jest.config`, `package.json`) → full suite
High fan-out	File imported by many modules — runs broader suite

Auto-escalation to full suite occurs when:

More than 70% of total tests are mapped by the diff (overhead not worth it)
Configuration files are modified (package.json, tsconfig.json, CI workflows, etc.) — config changes always trigger full suite

This keeps CI fast on small PRs while ensuring correctness on large changes.

When to Use

/ck:test - Run full test suite with coverage
/ck:fix [issue] - Fix failing tests automatically
Pre-commit/pre-push validation
CI/CD pipeline verification

Key Capabilities

Category	Tools	Coverage Target
Unit Tests	Jest, Vitest, pytest, cargo test, go test	80%+
Integration	API testing, DB interactions, service layers	Critical paths
E2E	Playwright, Cypress, Flutter integration tests	Critical paths
Coverage	Line, branch, function, statement analysis	80%+ overall
Build Check	TypeScript, linting, bundle size, compilation	100% pass

Common Use Cases

Solo Dev - Pre-commit validation

/ck:test

Validates all changes before commit with coverage report

QA Engineer - Bug regression prevention

/ck:test
Then review: Which areas have <80% coverage?

Team Lead - PR validation

/ck:test
Verify: All tests pass + coverage targets met

DevOps - CI/CD integration

/ck:test
Output: JSON report for pipeline integration

Full-Stack Dev - Multi-framework testing

/ck:test

Runs: Flutter analyze + Jest + pytest in sequence

Coverage-first approach: Always check coverage gaps after test runs - uncovered code = untested risk Fail fast: Use /ck:test before starting new work to catch environment issues early Fix atomically: Use /ck:fix [specific failure] for targeted repairs instead of batch fixes Performance baseline: Track test execution time - >60s indicates need for optimization Docker isolation: Run tests in containers for consistency across team environments

Debugger - Investigate test failures
Code Reviewer - Pre-test code quality
Fullstack Developer - Post-test build verification

Key Takeaway

Test execution isn’t about 100% coverage - it’s about 80%+ on critical paths with zero flaky tests. The tester agent finds failures, diagnoses root causes, and ensures your coverage targets before any commit goes through.