cohere-enterprise-rbac
Pass
Audited by Gen Agent Trust Hub on Sep 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill includes code templates for a chat interface that processes user-supplied messages. This creates a surface where malicious instructions in the data could potentially influence the LLM's behavior, although this is inherent to the primary purpose of building a chat wrapper.
- Ingestion points: The 'message' parameter in the 'teamChat' and 'budgetedChat' functions within SKILL.md.
- Boundary markers: Not included in the educational boilerplate code.
- Capability inventory: The code interacts with the 'cohere.chat' API endpoint.
- Sanitization: Not explicitly implemented in the provided architectural examples.
Audit Metadata