Incredible research, by an incredible team. The build-up from « uh there’s something here » to the full compromise was wild!
🚨👾First post → Awesome research.View quoted post on X
We compromised the latest version of OpenAI Codex.
We prompt injected.
We escaped a sandbox.
We exfiltrated secrets…
And ended up with control over what could be published to the official openai/codex npm package.
Prompt injection → RCE → supply chain compromise.
Full research by @0xMadder and @_Worty in the comments 👇