Crypto news

Ledger has introduced its new Agent Stack toolkit, designed to integrate AI agents with cryptocurrency wallets. This move marks a shift from simple asset storage to active management through smart algorithms, where hardware serves as the final arbiter of security.

Agent Stack has undergone closed testing with over 1,000 agents. The core concept behind this toolkit is simple and elegant: "agents propose, humans confirm, the Ledger device executes." This means AI assistants can check balances, analyze portfolios, and prepare transactions, but any operations involving fund movement require mandatory confirmation on the hardware wallet's screen. This approach ensures that even the most advanced algorithm cannot act without the user's knowledge.

What's included in Agent Stack

The toolkit includes several key components: Device Management Kit Skills, Ledger Wallet CLI, Ledger Enterprise CLI, and Ledger Enterprise Multisig CLI. According to the documentation, Ledger Wallet CLI allows an agent or user to retrieve account data, check balances and transaction history, and prepare asset transfers. Actions that do not require signing are performed without the device, speeding up routine processes, but critical operations remain protected by the hardware key.

Wallet CLI supports Bitcoin, Ethereum, and all EVM-compatible networks, as well as Solana. This covers major blockchain ecosystems, making the tool versatile for most users.

Expanding functionality: from crypto wallet to universal vault

Ledger is not limited to cryptocurrencies. Through OpenPGP, devices can store keys for accessing encrypted files, environment variables, and API tokens. And via FIDO2, they can act as hardware login keys for GitHub, npm, 1Password, Discord, and other services. This transforms Ledger into a universal security device for the user's entire digital space.

The company also unveiled its AI security roadmap for 2026. The second quarter is scheduled to launch identity and command-line tools, the third quarter will see Agent Intents and Agent Policies layers for confirmation and action rules, and the fourth quarter will introduce the Proof of Human mechanism. The latter is likely to be a revolutionary step in verifying human involvement in operations.

Context and conclusions

Earlier, Ledger Donjon researchers identified a vulnerability in Tangem hardware wallets related to laser fault injection. This serves as a reminder that even the most secure devices require constant auditing. However, Ledger itself, despite postponing its IPO due to market conditions, continues to actively develop its ecosystem, betting on security and AI integration.

My expert opinion: Agent Stack is not just another tool but a strategic step by Ledger toward creating a "security layer" for AI agents. In a world where algorithms are increasingly managing finances, hardware confirmation remains the only reliable way to prevent catastrophic errors or malicious actions. If Ledger successfully implements its roadmap, we may see a standard for the entire industry, where hardware wallets become not just a vault but a brain center for AI agents.