openai-docs
Warn
Audited by Snyk on Aug 19, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).
- Potentially malicious external URL detected (high risk: 0.90). The skill runs helper scripts and MCP tool calls at runtime that fetch and consume remote OpenAI docs (e.g., https://developers.openai.com/codex/codex-manual.md, https://developers.openai.com/api/docs/guides/latest-model.md) and contacts the MCP server https://developers.openai.com/mcp to drive prompt selection and model-migration logic, so these external URLs directly control agent behavior.
MEDIUM W013: Attempt to modify system services in skill instructions.
- Attempt to modify system services in skill instructions detected (high risk: 1.00). The skill explicitly instructs retrying an install command "with escalated permissions" (i.e., asking for sudo/privilege escalation) which directs the agent to obtain elevated rights and alter the machine state.
Issues (2)
W012
MEDIUMUnverifiable external dependency detected (runtime URL that controls agent).
W013
MEDIUMAttempt to modify system services in skill instructions.
Audit Metadata