codemap
Pass
Audited by Gen Agent Trust Hub on May 12, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [SAFE]: The skill provides structured templates and instructions for generating CodeMaps to help agents navigate codebases progressively.\n- [COMMAND_EXECUTION]: The skill utilizes
rg(ripgrep) to search local files, which is a common and appropriate tool for code analysis.\n- [DATA_EXFILTRATION]: No network access or data exfiltration mechanisms were found.\n- [INDIRECT_PROMPT_INJECTION]: The skill analyzes external code, presenting a standard surface for indirect prompt injection.\n - Ingestion points: Code files searched using
rginSKILL.md.\n - Boundary markers: Absent in the current templates.\n
- Capability inventory: Search via
rgand file writes tomydocs/codemap/.\n - Sanitization: Not explicitly implemented for the ingested code content.
Audit Metadata