@yuaone/core 0.9.0 → 0.9.2

package/dist/skills/skills/test-driven.md

@@ -1,51 +0,0 @@
-## Identity
-- domain: general
-- type: test
-- confidence: 0.85
-
-# Test-Driven — Red, Green, Refactor
-
-Write the failing test before writing any implementation. The test defines what "done" means.
-
-## Workflow
-1. **Red** — Write the test. Run it. It MUST fail. If it passes, the test is wrong.
-2. **Green** — Write the minimal code to make it pass. No extras.
-3. **Refactor** — Clean up without changing behavior. Tests must still pass.
-4. **Commit** — One test + implementation at a time.
-
-## Rules
-- Test behavior, not implementation. Test what it does, not how.
-- One assertion focus per test. Multiple assertions are fine if they all test the same thing.
-- Use real inputs and expected outputs — no magic numbers without explanation.
-- If you need a mock, you probably need to redesign the interface.
-
-## Known Error Patterns
-
-### Test Passes Without Implementation
-- **Symptom**: Wrote test, it passed immediately before any implementation
-- **Cause**: Test is testing the wrong thing, or the feature already exists
-- **Strategy**: 1. Read the test assertion carefully 2. Verify the thing being tested does not already exist 3. Make the assertion stricter until it fails
-- **Tool sequence**: file_read (test) → shell_exec (run test) → fix assertion
-- **Pitfall**: Do NOT skip the "verify red" step. A test that never fails is worthless.
-
-### Test Too Tightly Coupled to Implementation
-- **Symptom**: Refactoring breaks tests even though behavior is unchanged
-- **Cause**: Test asserts internal state, method call counts, or private methods
-- **Strategy**: Rewrite test to assert observable outputs and side effects only
-- **Tool sequence**: file_read (test) → file_edit (test only) → shell_exec
-- **Pitfall**: Do NOT test private methods directly.
-
-### All Tests Pass But Feature Is Incomplete
-- **Symptom**: Tests pass but edge cases are missing
-- **Cause**: Tests only cover the happy path
-- **Strategy**: 1. Identify edge cases: null/empty, boundary values, error conditions 2. Write one test per edge case 3. Implement
-- **Tool sequence**: file_read (implementation) → file_edit (add tests) → shell_exec
-- **Pitfall**: Do NOT consider a feature done with only happy-path tests.
-
-## Validation Checklist
-- [ ] Test written BEFORE implementation
-- [ ] Test runs RED before implementation
-- [ ] Test runs GREEN after implementation
-- [ ] Edge cases covered (null, empty, boundary, error)
-- [ ] No implementation details asserted
-- [ ] Full test suite passes