verifiable-execution

Build durable, autonomous, resilient and observable AI agents with built-in workflow orchestration, security, statefulness and telemetry.

Open protocol for AI agent work contracts and verifiable execution — authorization, evidence, and acceptance for multi-agent systems

Durable execution for Microsoft Agent Framework agents in .NET, on Dapr Workflow and Diagrid Catalyst

Open source shows what could run. Receipts prove what did run. A signed-receipt wrapper for the published xai-org/x-algorithm Phoenix demo pipeline. Apache-2.0.

verifier-as-service on x402. verify execution proofs with ERC-7710 permission on base mainnet.

Durable execution for TypeScript AI agents: Mastra and more, as Dapr Workflows on Diagrid Catalyst

Durable execution for Java AI agents: make Spring AI ChatClient calls, tool calls and chat memory crash-proof with Dapr Workflows on Diagrid Catalyst. Spring Boot starter, zero code changes, on Maven Central.

An open standard for verifiable execution provenance: a portable, third-party-verifiable record of what an agent did, in what order, and under what governance.

Dispatch work to any coding agent. Get a receipt.

A contract layer for verifiable hybrid AI execution — research pre-standard 0.36.0

Community DeepSeek Harness plugin for verifiable AI agent authorization, execution evidence, and offline delivery verification

ETHOnline Continuity build: policy-gated Agent-to-Agent commerce with The Graph provenance, Privy authorization, and Arc USDC settlement.

A runtime layer that compiles user intent into verifiable, auditable, rollback-safe execution chains — on top of an LLM.

Testnet lab binding one exact EVM call to a human-signed mandate: agent proposes, evaluator judges, isolated signer attests, vault executes the exact bytes or reverts

Durable execution for Go AI agents: LangChainGo and Eino graphs as Dapr Workflows on Diagrid Catalyst