Score uncommitted changes against codebase conventions into a markdown scorecard.
npx skills add https://github.com/jwadhwa2259/codescope --skill eval-jwadhwa2259
Please help me install this Agent Skill. Skill: eval Source: https://github.com/jwadhwa2259/codescope/tree/main/skills/eval Command: npx skills add https://github.com/jwadhwa2259/codescope --skill eval-jwadhwa2259
Score uncommitted changes against codebase conventions and generate a deterministic scorecard to help you assess risk before changes are merged.
Run /codescope:eval to score your current uncommitted changes.
High-intent search queries and answers about installing and using this skill.
You can score uncommitted changes by running a deterministic evaluation that parses your local diffs and generates a markdown scorecard grading conformity, risk, and import health before review.
Deterministic code scoring uses local data and predefined codebase conventions to quantify risk and conformity for uncommitted changes, producing a four-metric scorecard without relying on AI models.
Yes, the evaluation supports multiple modes including scoring current uncommitted changes, running a specified task and scoring the results, or benchmarking to produce a deterministic scorecard.
No, the scorecard is generated deterministically using local data and codebase conventions, meaning no AI models are required to quantify conformity, risk, or import health.
The scorecard features four equal-weighted metrics covering conformity, risk, and import health, which combine to produce a composite grade with actionable guidance for improvements.
Berries-WangSummarize uncommitted Git changes and identify potential risks.
CaptainCrouton89Automate software tests to verify functionality against acceptance criteria.
octavehqDiagnose and optimize Octave qualification agent scoring and routing logic.
Objective-ArtsReview and fix code against quality gates and security standards.
