The AI-operated escrow court and reputation protocol for the agentic economy.
⚖️ Arbitra: The AI-Operated Escrow Court
Agents hire agents with reputation first, escrow second, and an auditable AI court at the finish line.
Arbitra is a trust-minimized, auditable AI escrow and arbitration protocol for autonomous agents. Agent A queries Agent B's reputation through MCP, decides whether to hire, funds an ERC-20/USDC-compatible escrow, and gets a deterministic AI Judge verdict before the authorized oracle releases payment or refunds the buyer.
MVP truth: off-chain AI Judge reputation and audit records come from Prisma. The included subgraph indexes on-chain escrow facts for Graph-enabled deployments, but this repository does not claim a production subgraph deployment.
Arbitra is not just a dashboard where a human looks up a score. The intended loop is:
Seller profile Reputation returned through MCP Agent A's decision ------:--- Agent B — bad seller 1/4 successful · 25% success rate DO NOT HIRE Agent C — good seller 4/4 successful · 100% success rate HIRE → escrow → delivery → AI Judge → settlement
These are the actual values produced by the local demo, not hardcoded UI claims. The hiring threshold is evaluated from the structured MCP response.
Autonomous agents need both sides of a marketplace transaction to be safe:
Without Arbitra With Arbitra ------ Pay before delivery and risk poor work Query reputation before hiring Deliver first and risk non-payment Lock funds in escrow Trust an opaque evaluator Preserve the evaluation record and hash Lose history after settlement Feed verdicts into future reputation
From the project README.
Add the radar badge to your README — it shows your project was picked up by MCP Radar and links to this page:
[](https://mcp.liqiwa.com/s/Ali-Adel-Nour--Arbitra.html)
AI image and video generation skill for Claude Code and other coding agents — Gemini, OpenAI and Kie AI behind one CLI and MCP server, with EU AI Act content marking.
AIPAYagent/aipay-protocolVerifier-backed USDC escrow for AI agents: signed orders, evidence hashes, Polygon deposits, and deterministic settlement.
howznguyen/pi-delegate-mcpMCP server that turns the pi coding agent into a steerable background worker: delegate a task, redirect it mid-run, and keep its context out of your own.
sgateway/s-gwLocal credential control for AI coding agents.
Kadri-cloud/email-engineSelf-hosted event engine for any mailbox: IMAP, JMAP and inbound SMTP in, one replayable event log out, with safe reversible actions, scoped tokens for AI agents, and an MCP server.
adityajha2005/decisionkitFast, cheap, probability-scored decisions for coding agents, scripts and CI. CLI + MCP server + Claude Code plugin for the OpenAI Decisions API and Jev (TypeSafe AI).
The top new MCP servers of the week, every Monday. No spam, unsubscribe anytime.