to-spec
Pass
Audited by Gen Agent Trust Hub on Sep 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill defines a surface for indirect prompt injection by instructing the agent to ingest and analyze external content.
- Ingestion points: The skill directs the agent to "Read the full conversation and each source the user gives" and "Check code or docs for facts you can find."
- Boundary markers: The instructions lack explicit delimiters or instructions for the agent to treat external content strictly as data, which may lead the agent to follow malicious directives embedded in those sources.
- Capability inventory: The skill is primarily focused on text formatting and documentation; no high-risk capabilities such as network exfiltration, arbitrary command execution, or sensitive file access are utilized in the provided instructions or configuration.
- Sanitization: There are no requirements or logic provided for the agent to validate or sanitize the facts and requirements extracted from external sources.
Audit Metadata