code-review
Pass
Audited by Gen Agent Trust Hub on Oct 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied file content through the $ARGUMENTS variable, which creates a potential surface for indirect prompt injection attacks where malicious instructions inside the reviewed files could influence the agent's behavior.
- Ingestion points: The skill reads and analyzes file content passed via the $ARGUMENTS parameter in the SKILL.md task definition.
- Boundary markers: There are no explicit delimiters or instructions to the sub-agent to ignore natural language instructions found within the source code files.
- Capability inventory: The sub-agent has the capability to read file contents, perform logical analysis, and suggest code modifications (diffs).
- Sanitization: The skill lacks mechanisms to sanitize or filter out potential prompt injection patterns from the source code being reviewed.
Audit Metadata