mushi-debug
Pass
Audited by Gen Agent Trust Hub on Aug 7, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses local CLI tools including
mushiandsupabaseto perform diagnostic checks and retrieve system logs. These operations are standard for the described debugging use case. - [PROMPT_INJECTION]: The skill ingests data from external sources such as Sentry issue trackers and Supabase logs (Ingestion points). It does not employ specific boundary markers or sanitization for this external content. The capabilities within the agent's context are focused on project-specific CLI tools and MCP functions (Capability inventory). This design is consistent with the requirements for a debugging assistant.
- [CREDENTIALS_UNSAFE]: Instructions for API key setup and database migrations use standard placeholder formats (e.g.,
sk-ant-...) to avoid exposing sensitive information. The skill promotes secure CLI-based management of credentials.
Audit Metadata