codexforclaude

Pass

Audited by Gen Agent Trust Hub on Mar 14, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The documentation includes examples for adding MCP servers using npx from well-known providers such as Upstash (@upstash/context7-mcp) and the Model Context Protocol organization (@modelcontextprotocol/server-postgres).
  • [COMMAND_EXECUTION]: The guide describes the security model of the Codex CLI, detailing various sandbox modes and approval policies, including high-privilege options like danger-full-access and the --dangerously-bypass-approvals-and-sandbox flag for specific administrative or CI/CD use cases.
  • [PROMPT_INJECTION]: The skill explains how to use internal triggers like $skill-creator within the Codex environment to automate the creation of configuration files and skill structures.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 14, 2026, 04:34 AM