git-ai-archaeology
Pass
Audited by Gen Agent Trust Hub on Apr 15, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill processes untrusted data from git logs and CHANGELOG files which could contain malicious instructions.
- Ingestion points: Data from
git logandCHANGELOG.mdenters the agent context. - Boundary markers: None; the skill does not wrap external data in delimiters.
- Capability inventory: The agent utilizes
Write,Read, andBashtools. - Sanitization: No filtering or escaping of git data is performed before processing.
- [COMMAND_EXECUTION]: The skill uses the
Bashtool to executegitcommands with the{repo_path}placeholder. This pattern could lead to command injection if the repository path provided by the user is not properly sanitized by the platform.
Audit Metadata