agentsop-streaming-output
Pass
Audited by Gen Agent Trust Hub on Sep 4, 2026
Risk Level: SAFE
Full Analysis
- [REMOTE_CODE_EXECUTION]: The skill provides design patterns for integrating with official SDKs from trusted organizations including LangChain, OpenAI, and Anthropic. The implementation details rely on standard, well-documented API methods for streaming responses.
- [DATA_EXFILTRATION]: Guidelines describe the transmission of agent execution state and tokens over network protocols like Server-Sent Events (SSE). This is the intended functionality of the skill and follows industry-standard communication patterns.
- [INDIRECT_PROMPT_INJECTION]: While the skill involves processing LLM outputs that could be influenced by external data, it includes defensive guidance advising developers to validate or transform content before presenting it to users, specifically for safety or structural requirements.
- [SAFE]: The skill contains no malicious code, obfuscation, or unauthorized access patterns. It promotes security best practices such as implementing connection heartbeats to prevent silent failures behind proxies.
Audit Metadata