rhino-sdk-debug
Pass
Audited by Gen Agent Trust Hub on Mar 11, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: No security issues detected. The skill is composed entirely of markdown instructions designed to assist in debugging software errors. It does not interact with sensitive system files or perform network operations.
- [NO_CODE]: This skill contains no executable code or scripts. It relies on the AI agent to interpret the provided markdown guidance and reference documentation.
- [PROMPT_INJECTION]: The skill processes user-supplied error messages and code snippets, which represents a surface for indirect prompt injection. However, because the skill has no executable capabilities or tools, this surface cannot be exploited to perform malicious actions on the host environment.
- Ingestion points: User tracebacks and code snippets provided during debugging sessions.
- Boundary markers: Absent; there are no specific markers separating user data from system instructions.
- Capability inventory: None (no tools, subprocesses, or network capabilities are defined).
- Sanitization: None; the skill assumes the LLM's internal safety filters will handle malicious inputs.
Audit Metadata