/ 目錄 / 演練場 / SkillCompass
← 全部伺服器 ↗ GitHub
● 社群 Evol-ai ⚡ 即開即用

SkillCompass

作者 Evol-ai · Evol-ai/SkillCompass

Evaluate Agent Skill quality — find the weakest link, fix it, and prove the fix worked with before/after metrics.

SkillCompass scores your Agent Skills on clarity, activation rate, downstream correctness, and context cost. It highlights the skill most likely to be hurting your agent's performance, suggests a fix, and re-runs the evaluation so you can show the improvement. Useful when you have a shelf of skills and don't know which are actually earning their context weight.

▶ 觀看演示 📦 安裝 💡 使用場景

為什麼要用

核心特性

即時演示

實際使用效果

skillcompass-skill.replay ▶ 就緒
0/0

安裝

作者儲存庫 · Claude Skills 官方文件

使用場景

實戰用法： SkillCompass

Find the skill pulling your agent's performance down

👤 Skill authors with 5+ skills installed ⏱ ~45 min advanced

何時使用： You feel the agent has gotten worse, not better, as you added skills.

步驟
  1. Diagnose the loser
    For the weakest skill, what specifically is wrong?✓ 已複製
    → Concrete critique (vague description, conflicting with other skill, etc.)
  2. Propose a fix
    Suggest a minimal edit to SKILL.md to fix it.✓ 已複製
    → Small, reviewable diff
  3. Re-evaluate
    Re-run the eval and show before/after.✓ 已複製
    → Metrics improved, with evidence

結果： A measurably better skill bundle, with a reproducible eval process.

注意事項
  • Gaming the eval metric instead of helping real tasks — Include task-level downstream metrics (actual agent outcomes), not just text-level
搭配使用： skill-optimizer-skill · filesystem

Review a new skill before you publish it

👤 Skill authors shipping their first bundle ⏱ ~20 min intermediate

何時使用： Before pushing to GitHub and telling the world about your skill.

步驟
  1. Score the draft
    Evaluate my draft skill at ./my-skill/.✓ 已複製
    → Dimension scores
  2. Fix obvious issues
    Apply the low-hanging suggestions✓ 已複製
    → Edits in SKILL.md

結果： A publication-ready skill rather than a rough draft.

注意事項
  • Chasing a perfect score — Ship when scores plateau — diminishing returns

組合

與其他 MCP 搭配，撬動十倍槓桿

skillcompass-skill + skill-optimizer-skill

Two complementary tools: SkillCompass ranks, skill-optimizer drills into SKILL.md patterns

Use SkillCompass to pick the worst skill; use skill-optimizer to deeply analyze its SKILL.md.✓ 已複製
skillcompass-skill + filesystem

Operate across the full ~/.claude/skills/ directory

Evaluate every skill in ~/.claude/skills/ and give me a CSV.✓ 已複製

工具

此 MCP 暴露的能力

工具輸入參數何時呼叫成本
skill-scoring skill path(s) Periodic audits eval compute
weakest-link-id bundle scores After scoring 0
fix-suggestion weak skill + critique Before editing 0
before-after-eval pre/post SKILL.md After applying fixes eval compute

成本與限制

運行它的成本

API 配額
none beyond your LLM provider (evals use LLM calls)
每次呼叫 Token 數
evals can be heavy — budget 20–100k tokens for a full bundle scan
費用
free, MIT
提示
Run on one skill at a time during iteration; bundle runs only for audits

安全

權限、密鑰、影響範圍

憑證儲存： none at skill level
資料出站： none beyond your LLM provider

故障排查

常見錯誤與修復

Node errors on install

Ensure Node 20+; npm i inside the skill directory.

驗證： node -v
Evals are inconsistent run-to-run

Fix the task seed and use a non-stochastic sample; record provider+model.

替代方案

SkillCompass 對比其他方案

替代方案何時用它替代權衡
skill-optimizer-skillYou want a single skill analyzed deeply rather than a bundle rankedDepth over breadth
manual reviewYou have 1–2 skills totalDoesn't scale

更多

資源

📖 閱讀 GitHub 上的官方 README

🐙 查看未解決的 issue

🔍 瀏覽全部 400+ MCP 伺服器和 Skills