← ClaudeAtlas

comprehensive-ecosystem-auditlisted

Orchestrates all 8 ecosystem audits in staged waves with 4-concurrent-agent limit, aggregates results into a unified health report with cross-audit insights, domain heat maps, and priority-ranked findings.
jasonmichaelbell78-creator/sonash-v0 · ★ 2 · Web & Frontend · score 74
Install: claude install-skill jasonmichaelbell78-creator/sonash-v0
<!-- prettier-ignore-start --> **Document Version:** 1.1 **Last Updated:** 2026-02-24 **Status:** ACTIVE <!-- prettier-ignore-end --> # Comprehensive Ecosystem Audit Orchestrates all 8 ecosystem audits (hook, session, TDMS, PR, health, skill, doc, script) in 2 staged waves, aggregates results into a unified health report. **Invocation:** `/comprehensive-ecosystem-audit` **Output:** `COMPREHENSIVE_ECOSYSTEM_AUDIT_REPORT.md` in the project root. --- ## CRITICAL RULES 1. **CHECK for saved progress first** -- resume from `.claude/tmp/comprehensive-ecosystem-audit-progress.json` if < 2 hours old. 2. **Run audits via npm scripts / node commands**, not by reading checker code. 3. **CRITICAL RETURN PROTOCOL:** Each agent saves JSON to `.claude/tmp/ecosystem-{name}-result.json` and returns ONLY: `COMPLETE: {audit-name} grade {grade} score {score} errors {N} warnings {N} info {N}` 4. **Check completion via `wc -l`/`wc -c` on output files**, never read full output into context. 5. **Save progress after each stage** to survive context compaction. --- ## Compaction Guard State file: `.claude/tmp/comprehensive-ecosystem-audit-progress.json` Tracks stages (pending/completed), individual audit status (pending/running/completed/failed), and result summaries (grade, score, counts). | State | Resume Action | | ----------------------------- | ------------------------------------ | | No file or > 2 hours old | Start fresh