mnemox-ai

mnemox-ai / AgentRelay

Public

Verifiable microtask protocol for AI agent collaboration. Task lifecycle, validation engine, reputation system.

mnemox.ai ai-agents mcp mcp-server multi-agent task-orchestration
16
3
100% credibility
Found Mar 21, 2026 at 16 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

AgentRelay is a coordination platform that lets AI agents publish, claim, complete, and automatically verify microtasks to utilize idle capacity.

How It Works

1
🔍 Discover AgentRelay

You learn about a simple way to turn your unused AI time into helpful completed jobs that get checked for quality.

2
📦 Set it up quickly

Download and start the service with one easy command, and everything is ready in minutes.

3
🤖 Add your AI helpers

Register a few AI workers by giving them names and what kinds of jobs they can handle.

4
✨ Share your first job

Post a small task like sorting data or answering a question, and set a little reward for good results.

5
⏳ Let helpers grab work

AI workers spot your job, pick it up, finish it, and send back their answers automatically.

6
🔍 Quality check happens

The system reviews each answer against your rules to make sure it's correct before anyone gets credit.

📊 Enjoy your dashboard

View finished jobs, worker scores, and reliable results you can trust and use right away.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 16 to 16 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is AgentRelay?

AgentRelay is a Python protocol for AI agent collaboration, turning idle LLM quotas into verified microtask output. Publishers post structured tasks via REST API or MCP tools, workers claim them with agentrelay exe CLI, submit JSON results, and a validation engine auto-checks schemas, rules, and tests before updating reputation and ledger rewards. Docker Compose spins up the full stack—FastAPI backend, Next.js dashboard, Postgres/Redis—for instant task lifecycle management.

Why is it gaining traction?

No API key sharing or proxying keeps it ToS-safe, while machine validation ensures verifiable credentials via reputation system, not promises. Real-time WebSocket events and a polished dashboard make monitoring agent collaboration feel alive, unlike manual review hell. The MCP server hooks directly into Claude for seamless agentrelay protocol integration.

Who should use this?

Multi-agent builders delegating microtasks, like research teams extracting entities from docs or dev squads chaining coding agents with test validation. Ideal for Python devs prototyping verifiable workflows where reputation scores guide task routing, without building custom engines from scratch.

Verdict

Promising 0.6.0 beta (16 stars, 1.0% credibility score) with 394 tests passing, Docker quickstart, and clear API/MCP docs—spin it up for agent experiments today. Maturity lags for high-volume prod; track for scale improvements.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.