chain-llm-pattern

Pass

Audited by Gen Agent Trust Hub on May 11, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides an architectural pattern for building LLM chains that emphasizes validation and deterministic logic. No malicious behavior, data exfiltration, or obfuscation was detected.
  • [COMMAND_EXECUTION]: The provided n8n workflow (groq-chain-example.json) uses Code nodes to perform input validation and arithmetic. The JavaScript code is statically defined in the workflow and limited to data transformation tasks, posing no security risk to the host environment.
  • [EXTERNAL_DOWNLOADS]: The workflow utilizes standard n8n integrations and models from reputable providers (Groq, OpenAI, Claude). It does not download or execute untrusted third-party scripts or binaries.
  • [SAFE]: The skill addresses potential indirect prompt injection (Category 8) vulnerability surfaces by recommending strict schema binding via Information Extractor nodes and implementing validation checkpoints ('Validate Stage 1') that sanitize and verify LLM outputs before they influence downstream logic.
Audit Metadata
Risk Level
SAFE
Analyzed
May 11, 2026, 10:54 PM
Security Audit — agent-trust-hub — chain-llm-pattern