Raft 1.0 Enables Multi-Agent Team Collaboration
PrajwalTomar_ · x · 2026-07-16
Raft 1.0 has been released, focusing on enabling multiple AI agents to collaborate in a "team mode".
Author RC mentioned having built Kimi CLI at Moonshot and group chat bots in the past, and has been developing Raft publicly for the past four months. The goal is to solve common issues when working with multiple agents: scattered terminals, sessions, skills, and contexts, leaving humans to manually stitch the workflow together.
Product concepts include:
- Managing multiple agents within a single workspace
- Each agent having its own memory and expertise
- Assigning tasks across different models like Claude, Codex, and Hermes
- One agent can review and fix issues for another
- Agents can claim tasks in parallel and handle threaded handoffs
The quote emphasizes that humans are no longer constantly "prompt-driven" but act more like a conductor directing a team capable of memory and handoffs.
Related event: Raft 1.0 Launches Team Mode for Multi-Agent Workspaces(12 posts)→
More from coding & agent
- Poolside says Laguna S 2.1 is a 118B open-weight agentic coding model with 1M context — GavinSBaker · 2026-07-22
- Building a Knowledge Graph Without a Graph DB: 1000x Cheaper Than GraphRAG — TheRedfather · 2026-07-22
- FactoryAI gave back its first millions, then shipped Droid CLI two years later — matanSF · 2026-07-22
- Devin Outposts aims to run AI agents on any machine, from Mac minis to Kubernetes clusters — blaizedsouza · 2026-07-22
- Hermes Agent Refactoring Proposal: Decoupling via Event Bus and Monorepo Slicing — Promptmethus · 2026-07-22
- Show HN: a 6.2 MB pure-Go terminal command palette with no fzf dependency — MarinhoD · 2026-07-22