skills/jnpiyush/agentx/rag-pipelines/Gen Agent Trust Hub

rag-pipelines

Pass

Audited by Gen Agent Trust Hub on May 26, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill acts as a technical reference for building RAG pipelines, offering architectural advice and best practices without any executable malicious code.
  • [PROMPT_INJECTION]: The skill includes prompt templates designed for safety and grounding, specifically instructing the agent to use only provided context and to admit when information is missing, which mitigates standard injection risks.
  • [DATA_EXFILTRATION]: No sensitive data access or unauthorized network requests were found. All external URLs point to legitimate documentation for well-known developer tools and cloud services.
  • [REMOTE_CODE_EXECUTION]: The mention of scaffold-rag-pipeline.py refers to a local utility script for project initialization; no remote code downloads or untrusted execution patterns are present.
  • [COMMAND_EXECUTION]: The skill contains no commands that would lead to unauthorized system access or privilege escalation.
Audit Metadata
Risk Level
SAFE
Analyzed
May 26, 2026, 01:04 PM
Security Audit — agent-trust-hub — rag-pipelines