gc-review-a11y
Pass
Audited by Gen Agent Trust Hub on Mar 12, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection (Category 8).
- Ingestion points: The skill reads local source code files and git diff outputs in Step 1 and Step 3 to perform its review.
- Boundary markers: There are no explicit instructions or delimiters provided to the agent to disregard instructions embedded within the reviewed code (e.g., within HTML or JavaScript comments).
- Capability inventory: The skill uses an 'Edit' tool to modify files in Step 6 based on its findings, which could be abused if an injection is successful.
- Sanitization: No sanitization or filtering of the input data is performed before the content is processed by the LLM.
- [COMMAND_EXECUTION]: The skill executes local shell commands (
git rev-parse,git diff) to identify code changes for review. While these are necessary for its stated purpose, they represent a point of interaction with the local system environment.
Audit Metadata