OpenAI's Codex Security CLI and TypeScript SDK for finding, validating, and fixing security vulnerabilities. npm: https://www.npmjs.com/package/@openai/codex-security

Codex Security builds a threat model per repository, reproduces findings in a sandbox before showing them, and proposes patches a human must approve. The details.

Local-first workbench to run, inspect, compare, report, and gate OpenAI Codex Security scans.

Public sanitized defensive AI lab for threat modeling, Codex-assisted remediation, patch validation, and initial Codex Security evidence with human approval.

One broken AI agent a month: scanned with Codex Security, fixed with Codex, proven with tests. English and French.

Autonomous-agent operations & governance framework — fail-closed kill switch, cryptographic receipt chains, event-sourced state, A2A bus with idempotent fallback, sim-first delivery gates. Python, stdlib-only.