anthropic-managed-agents
Pass
Audited by Gen Agent Trust Hub on Sep 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill serves as a technical reference repository for the Anthropic Managed Agents API, containing documentation and code examples. All external URLs point to legitimate service domains (anthropic.com, claude.com, github.com) and all code snippets use official SDKs.
- [INDIRECT_PROMPT_INJECTION]: The skill documents the platform's surface for indirect prompt injection. Ingestion points: files, GitHub repositories, and memory stores. Capabilities: bash execution, file operations, and web search. Boundary markers and Sanitization: The documentation recommends 'read_only' access for memory stores to mitigate risks from untrusted input.
Audit Metadata