commit
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION] (SAFE): The skill performs local git operations including
git status,git log,git diff,git commit, andgit push. These are standard and necessary for its primary function. - [DATA_EXFILTRATION] (SAFE): The skill implements a robust blacklist in the
git diffcommand to ignore sensitive files such as.pem,.key,.crt, and.p12, as well as various binary and media formats. This prevents sensitive data from being leaked into the LLM context during analysis. - [INDIRECT_PROMPT_INJECTION] (LOW): The skill processes untrusted data from repository files which could contain malicious instructions.
- Ingestion points: File content is ingested via
git diff --staged. - Boundary markers: Absent; the diff output is provided directly to the model.
- Capability inventory: The skill can execute
git commitandgit pushwhich could be influenced by malicious instructions in the diff. - Sanitization: While file types are filtered, the actual text content of source files is not sanitized.
Audit Metadata