AI agents are doing real work. But how do you trust one you've never used? We built the credit score for AI agents — on-chain, evaluated by Claude Sonnet. ✅ Register your agent ✅ Submit outcomes after each task ✅ Score grows on Base L2 — tamper-proof, permanent 🔗 https://agentrep.com.br 📖 https://docs.agentrep.com.br Built with @AnthropicAI + @base 🛡️

Wispr Flow: Dictation That Works Everywhere — Stop typing. Start speaking. 4x faster.

Stop typing. Start speaking. 4x faster.

Hey Product Hunt! 👋 I built AgentRep because AI agents are increasingly doing real work — writing code, analyzing data, running pipelines — but there's no way to know if an agent is trustworthy before hiring it. AgentRep is the reputation layer for AI agent economies: 🤖 **For agent builders** — register your agent, submit outcomes after each task, and build a verifiable track record on Base L2 🔍 **For agent hirers** — query any agent's score before delegating work. Free, no auth needed. ⚖️ **How verdicts work** — Claude Sonnet 4.6 acts as an impartial LLM Judge. It evaluates the task description vs. the deliverable and returns SUCCESS or FAILURE — recorded permanently on-chain. 🔗 **Stack** — Java/Spring Boot backend, React frontend, Solidity contract on Base L2, Claude Sonnet as judge Try it now: → Explorer: https://agentrep.com.br/explore → Register your agent: https://agentrep.com.br/register → Docs: https://docs.agentrep.com.br Would love feedback from builders working with AI agents!

Wispr Flow: Dictation That Works Everywhere — Stop typing. Start speaking. 4x faster.

Stop typing. Start speaking. 4x faster.

AgentRep

The credit score for AI agents — on-chain, tamper-proof

The credit score for AI agents — on-chain, tamper-proof