From promptbook
Diagnose Promptbook setup — check config, hooks, API key, and session tracking health
How this skill is triggered — by the user, by Claude, or both
Slash command
/promptbook:doctorThis skill is limited to the following tools:
The summary Claude sees in its skill listing — used to decide when to auto-load this skill
Run diagnostics to check if Promptbook is set up correctly. Execute each check via Bash and report the results.
Run diagnostics to check if Promptbook is set up correctly. Execute each check via Bash and report the results.
Run these checks in order and report each result:
if [ -f "$HOME/.promptbook/config.json" ]; then
echo "CONFIG_FOUND"
# Check permissions (should be 600)
stat -f "%Lp" "$HOME/.promptbook/config.json" 2>/dev/null || stat -c "%a" "$HOME/.promptbook/config.json" 2>/dev/null
# Check required fields exist (without displaying values)
node -e "const c=JSON.parse(require('fs').readFileSync('$HOME/.promptbook/config.json','utf8'));console.log('has_api_key:',!!c.api_key);console.log('has_api_url:',!!c.api_url);console.log('auto_summary:',c.auto_summary);console.log('telemetry_consent:',c.telemetry_consent===true)"
else
echo "NO_CONFIG"
fi
If no config file found or telemetry_consent is false: tell the user to run /setup first.
Use the config file to test the API key without displaying it:
API_KEY=$(node -e "console.log(JSON.parse(require('fs').readFileSync('$HOME/.promptbook/config.json','utf8')).api_key)")
curl -sL -o /dev/null -w "%{http_code}" -X POST "https://promptbook.gg/api/auth/verify-setup" \
-H "Authorization: Bearer $API_KEY"
/setup againCheck for recent session data files to see if hooks are actually firing:
ls -lt ~/.promptbook/sessions/ 2>/dev/null | head -5
If no session files found: hooks may not be firing. Suggest starting a new Claude Code session.
Present results as a clear diagnostic:
Promptbook Doctor
─────────────────
✓ Config: Found (~/.promptbook/config.json)
✓ Consent: Granted during setup
✓ API Key: Valid and verified
✓ Activity: Last session 2 hours ago
Use ✓ for passing checks, ✗ for failures, ⚠ for warnings.
npx claudepluginhub promptbookgg/claude-code-plugin --plugin promptbookCreates, edits, and optimizes skills for Claude Code, including drafting, evaluating with test prompts, iterating on performance, and improving skill descriptions for better triggering accuracy.