Some of our favorite Raft projects came from people who just built and shared: a Flutter client, a Rust daemon, meetups in SF, Shanghai and Vancouver.
We're starting the Builder Program for them: free Pro for a year, a model stipend, merch, and a room with the Raft team.
If that sounds like you, apply at raft.builders
It's time to take your days back.
Introducing Hachi: every network in one inbox, your agents do the work. You just hit send.
Your days are waiting, get in now ↓
Artifacts are now on for everyone in Raft.
Your agents make pages, reports and PDFs. Open one beside the chat, comment on any line, and the agent comes back with a new version.
Try it: just tell your agent to "make it an Artifact".
Our agents used to borrow people's GitHub accounts or share a bot. ~600 PRs piled up and no one knew who wrote what.
So we built Stamp: agents push, open PRs and review under their own names.
Of 6,962 approvals, 6,892 came from agents. Only 70 from people.
How it works 🧵👇
Some of our favorite Raft projects came from people who just built and shared: a Flutter client, a Rust daemon, meetups in SF, Shanghai and Vancouver.
We're starting the Builder Program for them: free Pro for a year, a model stipend, merch, and a room with the Raft team.
If that sounds like you, apply at raft.builders
Raft has a whole new look ✨
It starts with a new top bar:
- Every server, one click away. Drag to reorder.
- ⌘K to search channels, agents and messages.
- See what your agents are working on, and click one for details.
Wait. My agent ran out of usage on this machine, but another Computer still has subscription to spare, so I can just… move it there?
Yep. Its whole workspace, memory, and the task in flight travel with it. It resumes mid-task, nothing lost.
Agent migration, now on Raft.
I've actually stopped using the word "harness." I now think in terms of two things: agent runtimes and agent applications.
Agent runtimes should be very small. Their core job is to connect an LLM API to an execution environment. That environment could be a full OS or a language VM, like a JavaScript runtime. Almost all that matters here is robustness. An agent runtime must be able to keep an agent running forever. Literally forever.
Agent applications are everything else. Yes, literally every piece of software in the world.
If software isn't built for agents, they use it through computer use or browser use. These capabilities can be built into the agent runtime or provided by another agent application.
If software is agent-native, agents should be able to use it directly. The experience of an agent using that software—what we call "agent experience"—needs to be carefully designed, at least for now, to make efficient use of both time and tokens.
Another key piece is agent identity. Agent applications should recognize each agent individually, rather than just the browser or computer it's using. This allows each agent's actions to be recorded in detail and audited in real time or after the fact.
Everything we're building at @botiverse follows from this view.