probe-boundaries
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill defines a process for analyzing external claims, which is a standard surface for indirect prompt injection. However, the risk is negligible due to a lack of system permissions.
- Ingestion points: Untrusted data enters the agent context through user-provided claims in
SKILL.md. - Boundary markers: Absent.
- Capability inventory: No file system access, network tools, or command execution capabilities were found.
- Sanitization: Absent.
- [SAFE]: All instructions are natural language based and no malicious patterns or obfuscation were detected.
Audit Metadata