openai-docs

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

• Potentially malicious external URL detected (high risk: 0.90). The skill runs local helper and resolver scripts and relies on an MCP server that fetch live docs at runtime (e.g., https://developers.openai.com/codex/codex-manual.md, https://developers.openai.com/api/docs/guides/latest-model.md, https://developers.openai.com/api/docs/guides/model-guidance?model=gpt-5.6 and https://developers.openai.com/mcp), and those fetched documents are parsed and used to determine model selection and prompting guidance, so the external content directly controls agent behavior.