netlify-ai-gateway

Pass

Audited by Gen Agent Trust Hub on Sep 9, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill describes workflows for processing external inputs (e.g., form submissions, chatbot queries) using AI models, which creates a surface for indirect prompt injection.\n
  • Ingestion points: netlify/functions/joke.js example and descriptions of summarizing form submissions in SKILL.md.\n
  • Boundary markers: No specific delimiters or "ignore" instructions are present in the code examples.\n
  • Capability inventory: The skill utilizes network capabilities via provider SDKs (OpenAI, Anthropic, Gemini, OpenRouter).\n
  • Sanitization: Explicit sanitization or filtering of external content is not demonstrated in the examples.\n- [EXTERNAL_DOWNLOADS]: The skill recommends installing official Node.js packages and development tools from trusted sources.\n
  • Evidence: Instructions include npm install for openai, @anthropic-ai/sdk, @google/genai, @openrouter/sdk, and @netlify/vite-plugin.\n
  • Evidence: Installation of the Netlify CLI via npm install -g netlify-cli@latest is documented for local development.\n- [COMMAND_EXECUTION]: The skill requires the use of shell commands for project initialization and deployment within the Netlify ecosystem.\n
  • Evidence: Commands such as netlify dev, netlify init, and netlify deploy --prod --open are utilized in the setup and development workflow.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 9, 2026, 03:52 AM
Security Audit — agent-trust-hub — netlify-ai-gateway