Skill Radar 📡
Your AI Skill Manager — Make every installed Skill count
Core Philosophy
Make Skill management simpler and more efficient.
- Simpler — See the full picture with one command, no more getting lost in 100+ Skills
- More Efficient — Precisely identify idle skills and gaps, maximize every investment
- Smarter — Data-driven decisions, not guesswork
Trigger Conditions
- "check skill", "skill radar", "skill diagnostics", "skill management", "skill optimization"
- "skill usage", "skill insight"
- "which skills am I not using", "should I install this skill", "do my skills need updating"
Features
| Command | Description |
|---|---|
usage |
📊 Usage Value Assessment (based on 5 data sources) |
status |
📋 Missing/Ready Status Check |
recommend |
💡 Smart Recommendations (discover capability gaps from conversation history) |
versions |
🔄 Version Check |
all |
Full Report (all commands above) |
Usage
# Full report (recommended for first use)
python3 <skill-path>/scripts/health_check.py all
# Individual checks
python3 <skill-path>/scripts/health_check.py usage # Usage value assessment
python3 <skill-path>/scripts/health_check.py status # Status check
python3 <skill-path>/scripts/health_check.py recommend # Smart recommendations
python3 <skill-path>/scripts/health_check.py versions # Version check
# Output as Markdown
python3 <skill-path>/scripts/health_check.py all > report.md
Usage Value Assessment
Integrates 5 data sources to infer each Skill's usage:
| Data Source | Content | Weight |
|---|---|---|
| Daily Memory | Events, decisions, progress | ⭐⭐⭐ |
| MEMORY.md | Core config, preferences | ⭐⭐⭐ |
| HEARTBEAT.md | Scheduled task config | ⭐⭐ |
| Session Logs | Raw conversation records | ⭐⭐ |
| AGENTS.md | Work rules | ⭐⭐ |
Scoring: 🔵 High (core tool) / 🟢 Medium / 🟡 Low / 🔴 Unused
Dependencies
- Python 3.8+ (pure standard library, no external dependencies)
- OpenClaw CLI (
openclaw) - ClawHub CLI (
npx clawhub, only for version check and recommendations)
