issue-update
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [DATA_EXPOSURE] (LOW): The skill uses hardcoded absolute paths (/Users/choegihwan/...), which exposes the local system's username and directory structure. This is a privacy concern and limits the skill's portability to other environments.
- [INDIRECT_PROMPT_INJECTION] (LOW): The skill is vulnerable to indirect injection as it processes data from git outputs and markdown files. 1. Ingestion points: git diff, git log, and local .md files. 2. Boundary markers: Absent. 3. Capability inventory: Local file writing and git command execution. 4. Sanitization: Absent.
- [COMMAND_EXECUTION] (SAFE): The skill invokes standard git commands (diff, log) to collect context. These are read-only operations appropriate for the skill's stated purpose.
Audit Metadata