Ledger Agent Stack: Secure AI Agents for Crypto

Ledger SAS has introduced Ledger Agent Stack, an open-source toolkit for developers. This toolkit helps create artificial intelligence agents that can read balances and draft crypto transactions. A key feature is that these agents cannot move funds without human authorization on hardware. The Ledger Agent Stack makes sure sensitive actions need confirmation on a Ledger hardware signer. Read-only requests, however, work on their own. This design protects funds and secrets even if an agent's software environment is compromised. The toolkit includes tools like the Device Management Kit Skills for integrating Ledger hardware with large language models. It also offers the Ledger Wallet CLI for programmatic access to crypto functions. For institutions, there's the Ledger Enterprise CLI and Ledger Enterprise Multisig CLI for hardware-anchored policy enforcement. New device apps also boost security beyond crypto, enabling OpenPGP encryption for agent secrets and gating access to services like GitHub and 1Password. The bottom line: This launch aims to reduce risks from human error and software vulnerabilities by anchoring critical signing decisions in hardware security modules.