We spent the last week rebuilding harness' main inference loop. Before it was quite static, question, answer, a couple of tool calls.
The core rewrite gives harness actual agentic capabilities. The ability to reflect, and decide, and use array of custom tools that differentiates harness from other browser assistants.
Some of the responses are even surprising us!
This is one of the few things that will be shipping with v2.
gHarness!
We spent the last week rebuilding harness' main inference loop. Before it was quite static, question, answer, a couple of tool calls.
The core rewrite gives harness actual agentic capabilities. The ability to reflect, and decide, and use array of custom tools that differentiates harness from other browser assistants.
Some of the responses are even surprising us!
This is one of the few things that will be shipping with v2.
gHarness!
The Starchild DAO Skill has been merged into @bankrbot skills repository!
You can now use Bankr to create and vote on proposals within Starchild DAO!
https://t.co/gkOLMXYvqH
Introducing Sakana Fugu: A full multi-agent orchestration system accessible via a single model API.
Our ‘Fugu Ultra’ model matches the performance of Fable and Mythos, delivering frontier capability without the risk of export controls.
Try it: https://t.co/hhO6qTawgb 🐡