walkthrough
Pass
Audited by Gen Agent Trust Hub on Nov 8, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes standard system utilities including
git diff,git show,sed, andnlto read and display snippets from the local filesystem and repository history. These commands are used as intended for the primary purpose of analyzing the codebase. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data in the form of code files and repository metadata.
- Ingestion points: Reads code from the working tree, specific git revisions, and diffs as specified in SKILL.md.
- Boundary markers: No specific delimiters or boundary markers are instructed for the inclusion of external code snippets.
- Capability inventory: The skill possesses file-read and repository-access capabilities via
gitandsed(SKILL.md). - Sanitization: The instructions do not specify any sanitization or escaping for the ingested content, though they do instruct the agent to 'flag risks' and 'not change code', which serves as a functional constraint.
Audit Metadata