Skill Health Dashboard

View and analyze skill health scores, trends, and optimization recommendations.

System Prompt

EXECUTION DIRECTIVE: When the user invokes this command, you MUST:

1. IMMEDIATELY execute - no questions, no explanations first
2. ALWAYS show full output from script execution
3. ALWAYS provide summary after execution completes

The user invoking the command IS the confirmation.

---

Usage

# Show dashboard summary
/skill-health

# View specific skill
/skill-health --skill codi-documentation-writer

# Filter by health status
/skill-health --min-score 70
/skill-health --max-score 50

# Filter by trend
/skill-health --trend improving
/skill-health --trend degrading

# Filter by category
/skill-health --category agents
/skill-health --category commands

# Compare to baseline
/skill-health --compare

# Record current scores
/skill-health --record

# Create baseline snapshot
/skill-health --baseline

Required Tools

Tool	Purpose	Required
Bash	Run skill-health-tracker.py script	Yes
Read	Access skill-health.json data	Optional

Script Dependencies:

• scripts/skill-health-tracker.py - Main health tracking
• context-storage/skill-health.json - Health data storage
• context-storage/baselines/ - Baseline snapshots

Execution

Run the skill health tracker:

cd ~/.coditect
source .venv/bin/activate

# Dashboard view (default)
python3 scripts/skill-health-tracker.py --dashboard

# Specific skill
python3 scripts/skill-health-tracker.py --skill "<skill-name>"

# Record current scores
python3 scripts/skill-health-tracker.py --record

# Create baseline
python3 scripts/skill-health-tracker.py --baseline

# Compare to baseline
python3 scripts/skill-health-tracker.py --compare

# Export full data
python3 scripts/skill-health-tracker.py --export --json

Dashboard Output

══════════════════════════════════════════════════════════════════════
                    SKILL HEALTH TRACKER
══════════════════════════════════════════════════════════════════════

📊 Overview: 150 skills tracked
   Average Health: 72.5%

   🟢 Excellent (90%+): 25
   🟡 Good (70-89%):    68
   🟠 Needs Work (50-69%): 42
   🔴 Critical (<50%):  15

📈 Trends:
   ↑ Improving: 23
   → Stable:    112
   ↓ Degrading: 15

🏆 Top Performers:
   orchestrator: 95% ↑
   codi-documentation-writer: 92% →
   senior-architect: 91% ↑

⚠️ Needs Attention:
   legacy-importer: 35% ↓
   old-migration-tool: 42% →

══════════════════════════════════════════════════════════════════════

Health Scoring

Score Components:

• Success Rate (50%): Successful invocations / total
• Trend Score (30%): Recent score trajectory
• Error Penalty (20%): Deduction for recurring errors

Status Levels:

Score	Status	Icon	Action
90-100	Excellent	🟢	Maintain
70-89	Good	🟡	Monitor
50-69	Needs Work	🟠	Improve
0-49	Critical	🔴	Urgent Fix

Trend Indicators:

Trend	Icon	Description
Improving	↑	Score rising >5%
Stable	→	Score within ±5%
Degrading	↓	Score falling >5%

Workflow

Daily Monitoring

# Quick health check
/skill-health

# Record after retrospective
/skill-health --record

Weekly Review

# Compare to baseline
/skill-health --compare

# Export report
/skill-health --export

When Issues Found

# View problematic skill
/skill-health --skill problematic-agent

# Check regressions
python3 hooks/skill-regression-detector.py --check

# Generate fix tasks
python3 hooks/skill-regression-detector.py --auto-task

---

Action Policy

<default_behavior> This command is read-only by default. Displays:

• Health score dashboard
• Trend analysis
• Top/bottom performers
• Recommendations

Use --record to update scores, --baseline to create snapshot. </default_behavior>

After execution, verify: - Dashboard displayed with all sections - Skill counts accurate - Trends calculated correctly - Recommendations relevant

Output Validation

Before completing, verify output contains:

• Overview with total skill count
• Average health score percentage
• Breakdown by status (Excellent/Good/Needs Work/Critical)
• Trend counts (Improving/Stable/Degrading)
• Top performers list (up to 5)
• Needs attention list (if any critical/low scores)
• Footer with available options

Success Output

When skill-health completes:

✅ COMMAND COMPLETE: /skill-health
Skills Tracked: N
Average Health: XX%
Critical: N skills need attention
Trend: N improving, N degrading

Completion Checklist

Before marking complete:

• Script executed successfully
• Dashboard displayed
• Health scores shown
• Trends calculated
• Recommendations provided (if applicable)

Failure Indicators

This command has FAILED if:

• ❌ Script not found
• ❌ Database connection error
• ❌ No health data available
• ❌ Dashboard not displayed

When NOT to Use

Do NOT use when:

• Need to fix a skill (use remediation directly)
• Looking for specific component (use /which)
• Running retrospective (use /retrospective)

Anti-Patterns (Avoid)

Anti-Pattern	Problem	Solution
Ignore critical skills	Issues persist	Address <50% skills
Skip recording	Stale data	Run --record after sessions
No baseline	Can't detect regression	Create with --baseline

Related Commands

Command	Purpose
/which <task>	Find best agent (now shows health)
/retrospective	Run session analysis
/optimize-skills	Alias for this command

Principles

This command embodies:

• #9 Based on Facts - Score-based health tracking
• #6 Clear, Understandable - Visual dashboard format
• #3 Complete Execution - Full health overview

Full Standard: CODITECT-STANDARD-AUTOMATION.md