AgentMint – Cryptographic proof of human approval for AI agent actions
AgentMint – Cryptographic proof of human approval for AI agent actions
Session auth says "agent can use Stripe." It doesn't say "Alice approved this $50 refund at 10:03am." AgentMint signs short-lived tokens proving a human approved a specific action. Ed25519 signatures Single-use (JTI tracking) 60s TTL default 3ms verify Full audit trail Built in Rust. Drop-in sidecar for your backend. https://github.com/aniketh-maddipati/agentmint
Share cardActual performance
Launch Intel predictions
Analyze your own launch →Correct prediction on native model
Similar products
NotaryOS – Cryptographic proof of what your AI agent didn't do
Approve your AI agent's actions on the go
Proof of Human API
Arc – Delegated authority and audit for AI agent actions
Nobulex – Cryptographic receipts for AI agent actions
Get a Merkle-Proof from Ethereum
Burni – Proof-of-Burn Ethereum NFTs
Proof of God
I implemented gmaxwell's "solvency proof scheme"
FLP Impossibility Theorem Proof in Coq