bmc-dashboard-debug-reviewer
Reviews Audit Runner output, extracts issues and logs, exports to structured files, and produces DEBUG-REPORT.md. Runs AFTER bmc-dashboard-audit-runner. Use when user asks for bmc debug review, export
SKILL.md.cursor/skills/bmc-dashboard-debug-reviewer/SKILL.md
GitHubBMC Dashboard Debug Reviewer Skill
Coordinates the Debug Reviewer Agent to analyze the audit output and export logs/issues.
When to Use
| Trigger | Action |
|---|---|
| "debug review" | Run after audit |
| "export logs and issues" | Same |
| "review bmc issues" | Same |
| After "AUDIT RUNNER COMPLETE" | Run Debug Reviewer |
Prerequisites
- Sequential: Audit Runner must complete first
- Input:
.cursor/bmc-audit/latest-report.md,handoff.json
Propagation
Si los hallazgos afectan a Design, Networks o Mapping: actualizar docs/team/PROJECT-STATE.md y consultar tabla de propagación en docs/team/PROJECT-TEAM-FULL-COVERAGE.md §4.
Execution Checklist
- Read handoff.json and latest-report.md
- Parse for errors, warnings, anomalies, config gaps
- Export to
.cursor/bmc-audit/debug-export/:- issues.md (by severity)
- logs-raw.txt
- config-gaps.md
- recommendations.md
- Write DEBUG-REPORT.md
- Emit "DEBUG REVIEWER COMPLETE. Output: .cursor/bmc-audit/"
Severity Levels
| Level | Examples |
|---|---|
| Critical | API down, service-account missing, .env missing |
| High | 503 on cotizaciones, auth failures, Sheets not configured |
| Medium | npm audit findings, ngrok not running |
| Low | No persistent logs, minor config suggestions |
Output Structure
.cursor/bmc-audit/
├── latest-report.md (from Audit Runner)
├── handoff.json (from Audit Runner)
├── DEBUG-REPORT.md (from Debug Reviewer)
└── debug-export/
├── issues.md
├── logs-raw.txt
├── config-gaps.md
└── recommendations.md
Reference
- bmc-dashboard-audit-runner — upstream agent