openai-docs

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 uses the mcp__openaiDeveloperDocs__search_openai_docs / fetch_openai_doc tools to ingest OpenAI developer documentation text from the MCP server (an external, third-party authored source), which is then placed into the agent’s LLM context.

MEDIUM W013: Attempt to modify system services in skill instructions.

• Attempt to modify system services in skill instructions detected (high risk: 0.90). The skill tells the agent to run system-level install commands and explicitly instructs retrying with "escalated permissions" if the MCP install fails, which encourages privilege escalation and modification of the machine's state.