rmslop
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external data in the form of code diffs against a development branch.
- Ingestion points: The agent reads file content and git diffs (SKILL.md).
- Boundary markers: Absent. There are no explicit instructions to ignore instructions embedded within the code being reviewed.
- Capability inventory: The agent uses file system tools to read and modify source code.
- Sanitization: Absent. The agent processes code changes directly.
- This represents a standard surface for indirect prompt injection inherent to code maintenance tasks, where malicious instructions hidden in a diff could influence the agent.
Audit Metadata