Claude Code skill that enforces task scope boundaries during agent execution, preventing drift beyond the approved plan.
Stars
3
Forks
1
Watchers
3
Open Issues
0
Overall repository health assessment
No package.json found
This might not be a Node.js project
11
commits
feat(skill): SKILL.md v5 with anti-rationalization defense and hardened eval harness (#8)
838821eView on GitHubfeat: motorcycle tier eval — enriched context + LLM-as-judge (#6)
92a825dView on GitHubfeat: SKILL.md accuracy improvements and multi-run harness (#5)
77ff42aView on GitHubdocs: add full README with installation, usage, and SCOPE.md example
7bc361fView on GitHubtest: document scope-lock GREEN phase results — all 4 scenarios pass
18cc6acView on GitHubfix: correct description prefix and add Logging Rules heading in SKILL.md
91f6ec1View on GitHubfix: add Red Flags section and fix table structure in scope-lock skill
211b33aView on GitHubfeat: add scope-lock SKILL.md (GREEN phase)
f124489View on GitHub