Ledger launches Agent Stack toolkit to enable AI agents to interact with crypto wallets while keeping keys secure

On July 16, 2026, Ledger launched Ledger Agent Stack, an open-source toolkit enabling AI agents to interact with crypto wallets while keeping private keys secure on hardware devices. The toolkit allows AI to draft transactions but requires physical human approval for execution. This launch, part of Ledger's 2026 AI roadmap, addresses security risks, as studies show 26.1% of agent skills have vulnerabilities, while 60% of breaches stem from human error. Partners like MoonPay have already integrated the tool.

Ledger Agent Stack launch

• ▪Ledger Agent Stack represents the first product release under Ledger's 2026 AI roadmap, which includes future tools for agent identity and proof of human checks.
• ▪Ledger launched Ledger Agent Stack, an open-source toolkit allowing artificial intelligence agents to interact with crypto wallets, on July 16, 2026.
• ▪MoonPay and Shisa.ai have integrated Ledger Agent Stack support to allow Ledger hardware connectivity within their respective products.

AI agent security architecture

• ▪The Ledger Agent Stack architecture prevents unauthorized fund transfers and credential access even if the artificial intelligence agent's software environment is compromised.
• ▪Ledger Agent Stack allows artificial intelligence agents to read balances and draft transactions while requiring physical human sign-off on hardware before execution.

Hardware wallet integration features

• ▪The Ledger Wallet CLI provides artificial intelligence agents with programmatic access to check balances, receive funds, and prepare sends, swaps, and staking.
• ▪Ledger Agent Stack includes Device Management Kit Skills, which are open-source instructions teaching large language model runtimes to integrate Ledger hardware.
• ▪Ledger integrated OpenPGP and Security Key features, enabling users to encrypt agent secrets and use Ledger devices as physical security keys for external services.

Enterprise treasury management tools

• ▪Ledger Agent Stack enables artificial intelligence agents to interact with both personal and institutional crypto wallets while keeping transaction approvals locked behind hardware.
• ▪The Ledger Enterprise CLI and Ledger Enterprise Multisig CLI extend hardware-anchored policy controls and security module enforcement to institutional treasury teams.

AI agent vulnerability statistics

• ▪Research cited by Ledger indicates that approximately 26.1% of all artificial intelligence agent skills contain at least one security vulnerability.
• ▪Studies cited by Ledger attribute approximately 60% of all security breaches to human error.

3 sources