code-review
Pass
Audited by Gen Agent Trust Hub on Sep 1, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted repository content which can serve as a vector for indirect prompt injection. Ingestion points include commit logs, git diffs, and documentation files (SKILL.md, sections 1 and 2). The instructions do not define delimiters to isolate untrusted content from the agent instructions, and no sanitization of this content is specified.
- [COMMAND_EXECUTION]: User-provided references are used directly in git commands, potentially exposing a command injection surface if the environment is not secured. The skill uses the variable in git rev-parse and git diff commands (SKILL.md, section 1). A validation step using git rev-parse is included as a mitigation.
Audit Metadata