health
Pass
Audited by Gen Agent Trust Hub on May 21, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes a series of local bash and Python scripts (e.g., collect-data.sh, check-maintainability.sh, check-agent-context.sh) to gather repository metrics and verify configuration integrity.
- [EXTERNAL_DOWNLOADS]: References the tw93/Waza repository via npx to resolve script paths; this is a legitimate vendor resource associated with the skill's author.
- [PROMPT_INJECTION]: The skill audits untrusted data from the local repository (such as project instructions and conversation logs), representing an indirect prompt injection surface. This is mitigated by explicit instructions to sub-agents to disregard any instructions embedded within the processed content.
- [DATA_EXFILTRATION]: Accesses sensitive local files such as ~/.claude/settings.local.json and conversation history for the purpose of the audit; however, no evidence of network transmission was found, and the skill includes logic to redact sensitive strings before processing.
Audit Metadata