Test driven execution
Skill markoblogo/abvx-agent-skills/skills/test-driven-execution
Reviewable capability layer for coding agents. Portable skills, delivery gates, workflow patterns, and verification-first engineering.
npx -y skills add markoblogo/abvx-agent-skills --skill test-driven-executionAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 4 stars4 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its author says it does
Copied from the file, not written here
Execute feature work and bug fixes with a tracer-bullet red-green-refactor loop. Use when building behavior that should be guided by tests, when fixing regressions, or when the user wants TDD or test-first execution. Prefer public interfaces, vertical slices, and behavior-focused tests that survive refactors.
The file declares its own license as MIT. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.
SKILL.md
1.9 KB, as published. Nobody here has run it
Test Driven Execution
Use tests as the feedback loop that shapes the implementation one vertical slice at a time.
Core Rules
- Test behavior through public interfaces.
- Prefer integration-style tests over implementation-coupled tests.
- Work in vertical slices, not horizontal batches.
- One failing test -> one minimal implementation -> repeat.
- Refactor only while green.
Workflow
- Confirm the behavior to build or fix.
- Identify the highest useful seam to test.
- Write one failing test for one observable behavior.
- Implement the minimum code to make it pass.
- Repeat for the next behavior.
- Refactor only after the slice is green.
For a diagnosed regression that needs durable proof, capture the exact same targeted command before and after with bug-evidence-protocol, then capture relevant broader checks. Do not infer FIX_PROVEN from a green targeted test alone.
Anti-Pattern
Do not write all tests first and all code later. That produces test suites for imagined behavior instead of validated behavior.
Good Tests
- name the user-visible behavior;
- survive internal refactors;
- avoid private methods and internal collaborator trivia;
- verify what the system does, not how it does it.
Planning Questions
- Which behaviors matter most?
- Which seams already exist?
- What should not be tested at this layer?
- What prior art already exists in the repo?
Final Report
Include the behaviors covered, the seams used, what remains untested, and any refactor opportunities exposed by the loop.