maintain
Pass
Audited by Gen Agent Trust Hub on Apr 6, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: Orchestrates various shell commands to perform codebase analysis, including linting, type checking, and git branch management. These operations are restricted to standard development tools and are essential for the skill's stated purpose.- [EXTERNAL_DOWNLOADS]: References the 'bunx' utility to run maintenance packages and directs users to the Linear CLI repository on GitHub for optional integration. These references target established development tools and do not involve automated execution of untrusted remote scripts.- [PROMPT_INJECTION]: Parses project-specific data such as source code comments and package metadata to identify markers like 'TODO' or 'FIXME'. This ingestion surface is used solely for generating health reports and suggesting fixes, with limited capability for influencing agent behavior due to the structured nature of the report.
Audit Metadata