News · DiggingBeagle record

PACE moves wallet-agent authorization outside the LLM

PACE treats the wallet agent as potentially compromised and moves authorization into a deterministic verifier plus smart-account enforcement. It reported zero unsafe executions in 2,800 benchmark trials.

A dated report connected to the underlying research where available.

The report

PACE starts from a useful security assumption: the LLM controlling a wallet may be wrong, manipulated or compromised. Instead of asking the model to obey policy, the system converts proposed actions into typed intents and checks them with deterministic policy logic.

Approved actions are bound to a signed Policy Decision Record and exact execution bytes, with expiration and replay protection enforced by the smart account. Across 2,800 deterministic benchmark trials, the authors reported zero unsafe executions.

That is logic-level benchmark evidence rather than production proof, but the architecture is important because it moves authorization outside the model.

Research behind this

Cite this record

DiggingBeagle. “PACE moves wallet-agent authorization outside the LLM.” https://diggingbeagle.com/news/pace-policy-attested-contract-execution-august-2026/

Citation guidance

Why this archive exists

The source matters after the headline fades.

DiggingBeagle is a non profit research project documenting AI security incidents, agent failures, vulnerabilities and AI-assisted operations. A case keeps its claims beside the sources that support, contest or limit them. Later updates stay visible, so a reader can see when the account changed.

We publish case reconstructions, dated reporting and analysis across records. Each has a different evidentiary role. About the project and our methodology explain how the work is reviewed.