Plugin Claude Code
Local-only skill quality and security evaluator for Claude Code/OpenClaw. Scores six dimensions, runs local validators, and stores reports and snapshots under .skill-compass/.
Instructions file CodexOpenCode
Instructions for Evol-ai/SkillCompass, covering agent instructions, required rules and source of truth.
Skill Claude CodeCodex
Evaluate skill quality, find the weakest dimension, and apply directed improvements. Also tracks usage to spot idle or risky skills. Use when: first session after install, or user asks about skill quality, evaluation, inbox, suggestions, or improvement.
Command
Use the Glob tool to find all /SKILL.md files recursively under the specified directory. Also check /.claude/skills/ if scanning project-level.
Command
Command "eval-compare" from Evol-ai/SkillCompass, covering /eval-compare — version comparison, arguments, steps, step 1: resolve versions and step 2: check cached results.
Command
Command "eval-evolve" from Evol-ai/SkillCompass, covering arguments, prerequisites, what this command does, step 1: validate and step 2: read current state.
Command
If --internal is passed, skip all interactive prompts — return results silently. This flag is set when eval-improve is called by other commands (e.g., from setup or skill-inbox action queue).
Command
Use the Read tool to load .skill-compass/{skill-name}/manifest.json. Verify.
Command
Use the Read tool to load .skill-compass/{skill-name}/manifest.json.
Command
Parse arguments. Use the Read tool to load the target SKILL.md file.
Command
🚀 Enhanced with Local Validators: This command now uses local JavaScript validators for D1, D2, and D3 dimensions to significantly reduce token consumption while maintaining evaluation quality. Complex reasoning tasks (D4, D5, D6) continue to use LLM evaluation with local pre-analysis.
Command
Triggered by the SessionStart hook (hooks/scripts/session-tracker.js), which compares the current SkillCompass version against the last recorded version. On first install, reinstall, or version change the hook injects a context message instructing Claude to load this file and follow it on the user's first interaction.
Command
This command gives users a quick local inventory of installed skills and surfaces only high-signal issues. It supports two modes.
Command
This command accepts free-form natural language and routes to the appropriate SkillCompass command. Also accessible as /skill-compass.
Command
Unified entry point for managing skill suggestions and browsing all installed skills. Provides two views: suggestions (default) and all skills.
Command
Generate a comprehensive report of all installed skills: quick health scan, context budget, portfolio overview, and quality summary.
Command
Check installed skills for remote updates and guide the user through updating and evaluating them.
Hook
Runs after a tool call finishes for Skill, Write and Edit tool calls, executing skill-usage-tracker.js, post-skill-edit.js and eval-gate.js via node (3 commands). From Evol-ai/SkillCompass.
Hook
Runs when a session starts, executing session-tracker.js via node. From Evol-ai/SkillCompass.
Hook
Runs when a session ends, executing session-tracker.js via node. From Evol-ai/SkillCompass.