# mothBot

- **Event:** [WeaveHacks 3: Self-Improving Agents Hackathon with Weights & Biases](https://cerebralvalley.ai/e/weave-hacks-3-self-improving-agents-hackathon-with-weights-and-biases-7014fe80)
- **When:** Jan 31 at 9:00 AM – Feb 1 at 5:00 PM (PST)
- **Where:** 400 California St, San Francisco, CA
- **Team:** [vivek shukla](https://cerebralvalley.ai/u/weebhek)
- **GitHub:** https://github.com/vivek100/mothBot
- **Demo video:** https://youtu.be/Bx997YR7Ov0
- **Gallery:** https://cerebralvalley.ai/e/weave-hacks-3-self-improving-agents-hackathon-with-weights-and-biases-7014fe80/hackathon/gallery
- **Page:** https://cerebralvalley.ai/e/weave-hacks-3-self-improving-agents-hackathon-with-weights-and-biases-7014fe80/hackathon/gallery/40

Mothbot is a selfevolving agent which maintains a space ship. It can call tools in chain instea dof running diagnostic on hull, then oxygen system and then gravity check, it cna just triggern them all in a chain in one tool call. We use marimo to run the tool chain engine its kind of like notebook, and then agent can save these tool chains as skills with descriptoin. So next time when you want to run diagnostics on hull, oxygen and gravity systems you can just reuse this, this also allows agent to fix the tool chains as agent knows what each tool is.

We use weaves to run tracing , and then we run evals at thread levels to add skills and then we do eval with coding agent to update prompts. This creates long running evolving loop.

We use pipekit for voice and agent orchestration, marimo for tool chain engine and then weavs for tracing and evals.

---

Markdown version of https://cerebralvalley.ai/e/weave-hacks-3-self-improving-agents-hackathon-with-weights-and-biases-7014fe80/hackathon/gallery/40. Site index for agents: https://cerebralvalley.ai/llms.txt · full text: https://cerebralvalley.ai/llms-full.txt