tipkit
Pass
Audited by Gen Agent Trust Hub on May 12, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill defines a procedure for the agent to fetch documentation from a non-whitelisted third-party domain (sosumi.ai).
- [INDIRECT_PROMPT_INJECTION]: Processing remote third-party documentation at runtime presents a surface for indirect prompt injection. 1. Ingestion points: External Markdown from sosumi.ai. 2. Boundary markers: None identified. 3. Capability inventory: Active agent reasoning and tool-use. 4. Sanitization: No content validation or sanitization specified.
Audit Metadata