AgentClash (@AgentClashDev) header
AgentClash (@AgentClashDev) profile picture

AgentClash

@AgentClashDev

https://teapaw.org/t.co/hE3jnC5Dos | Open Source AI Agent Evaluation Platform

Joined May 2026
  • Tweets24
  • Following12
  • Followers24
We just shipped Fleet Run evals at scale in the cloud AWS, Azure, Vercel, you name the cloud, we support it. One YAML → every pack × agent × model × repeat, racing in parallel sandboxes, scored live on a matrix They test prompts. We eval agents Try it: github.com/agentclash/agentc… npm i -g agentclash
6
5
1
13
1,590
AgentClash retweeted
Atharva (@attharrva15) profile picture
Atharva
@attharrva15Jul 1
1/ Check out Datasmith, a new primitive in @AgentClashDev! A clean, provider-agnostic Python SDK for generating high-quality synthetic datasets using a weak-vs-strong agentic loop.Inspired by Meta FAIR’s Autodata work. Built for people actually shipping agents and fine-tunes.
2
3
5
250
Try agentclash for free guys! agentclash.dev/tryouts
Image from AgentClash (@AgentClashDev)
8
10
20
1,484
guys we launched something proofswe.com/
16
8
46
5,819
about to launch*
35
AgentClash retweeted
Atharva (@attharrva15) profile picture
Atharva
@attharrva15Jun 19
The Next Unit of AI Evaluation Is the Session, not Benchmarks.

The Next Unit of AI Evaluation Is the Session, not Benchmarks.

Article · @attharrva15
1
1
1
4
461
I m verified guys
1
33
We just shipped some more cool stuff! agentclash.dev/changelog/202…
May 25 – Jun 01, 2026 — AgentClash Changelog

May 25 – Jun 01, 2026 — AgentClash Changelog

The datasets platform shipped end-to-end — import, generation, trace ingest, eval gates, and full workspace UI. /try CLI demos, multi-turn transcript exports, Hermes harness support, and PostHog...

agentclash.dev
1
40
we are about to change how agents are made
3
37
Load more