ai-rag-pipeline
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill implements a workflow that fetches content from external web search tools (Tavily and Exa) and interpolates it directly into LLM prompts. This creates a surface for indirect prompt injection, where malicious instructions embedded in retrieved web pages could attempt to influence the agent's behavior.
- [COMMAND_EXECUTION]: The skill heavily utilizes the
beltCLI tool for core operations, including user authentication (belt login) and executing remote AI applications (belt app run). These operations are consistent with the skill's stated purpose of managing RAG pipelines. - [EXTERNAL_DOWNLOADS]: The documentation references external installation scripts and resources hosted on the
inference-shGitHub repository. These are recognized as vendor-owned resources for thebeltecosystem.
Audit Metadata