using-neon

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.85). The skill’s runtime workflow fetches Neon documentation pages from https://neon.com/docs/... (public web content) via curl, and that fetched markdown becomes LLM-readable context, which is an outsider-authored free-text source.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 1.00). The skill explicitly instructs the agent to fetch documentation at runtime (e.g., curl https://neon.com/llms.txt and curl -H "Accept: text/markdown" https://neon.com/docs/), meaning externally-hosted content would be loaded during execution and could directly control prompts/responses.