Drop into a skill folder
Personal: ~/.claude/skills/ai-code-scaffold-test-loop/SKILL.md
Project: .claude/skills/ai-code-scaffold-test-loop/SKILL.md
Brali Agent Skill · reviewed · usable
Use AI to produce a first implementation for a bounded coding task, then inspect the diff, run tests, add missing edge cases, and reject code you cannot explain. A controlled experiment found faster completion on one coding task; it did not establish universal software quality gains.
Portable instruction
Define the behavior and constraints first. Ask AI for a small implementation. Read the diff, run the existing test suite, add tests for edge cases and failure paths, and inspect security or data-handling implications. Keep only code you can explain and maintain.
Did AI shorten implementation time without increasing code I do not understand, test, or want to own?
Install
Inspect SKILL.md first. The evidence state and skill mode are part of the contract.
Install for normal use.
Personal: ~/.claude/skills/ai-code-scaffold-test-loop/SKILL.md
Project: .claude/skills/ai-code-scaffold-test-loop/SKILL.md
hermes skills install https://brali-lifeos.github.io/skill-packs/ai-code-scaffold-test-loop/SKILL.mdWorkspace: <workspace>/skills/ai-code-scaffold-test-loop/SKILL.md
Global: ~/.openclaw/skills/ai-code-scaffold-test-loop/SKILL.md
Brali verification
Canonical ID: brali:ai-code-scaffold-test-loop
Evidence state: reviewed
Skill mode: usable
Recommendation eligible: yes
SHA-256: 5cbb1c9193cccbee495bed3b6b78d80d3601134a5312870339c2e6646cae7f91
Brali CI verified. This means identity, state, mode, provenance and file integrity matched the generated corpus. It is not third-party vendor endorsement.
Brali keeps a one-to-one machine surface: one hack record, one stable skill identity. Trust state determines what the skill is allowed to do.
Original Brali knowledge is available for non-commercial reuse under CC BY-NC-SA 4.0. Keep attribution, canonical record link, evidence state and skill mode.