Most security scanners run your private code through someone's black box. @odin_scan runs it in an Intel TDX enclave and hands you the attestation to prove it! Odin Scan's AI analysis runs on @near_ai's private inference, inside a hardware-sealed Intel TDX enclave. Your code is decrypted only inside it, invisible to the host, the infrastructure operators, and the model provider alike! So the report ships the proof: the raw Intel TDX quotes for the gateway and the model that served your scan, the nonce that binds them to that specific report, and a link to Intel's attestation roots. Open the report, expand "Confidential inference verified" and check it yourself. The same evidence is included in the PDF as well. Don't trust. Verify 🔥

Today @odin_scan moved all its inference to @near_ai private TEEs! 🔥 Your code and your prompts are now encrypted end to end. Not even NEAR can read them, and you can verify it with a cryptographic attestation. There is no bigger upgrade for a security scanner than that. 👇

When I started building @odin_scan at first, I thought running 3 models in parallel was going to be better than running 1. At that time it made sense, but it added a lot of complexity. Today, that was the accuracy ceiling in @odin_scan, which is now fixed! One agent that keeps full context and attacks its own findings beat it, on the same model: F1 0.89 → 0.97 17/17 bugs found 3 false positives → 1 @odin_scan is very good at findings smart contract exploits today! Check it out 👇 odinscan.ai/blog/odinscan-ac…

Ran @MANTRA_Chain's mantra-dex through @odin_scan. 91! Zero Critical, zero High. The three mediums are worth a look but the foundation is clean 🚀