LithosAI

Open-source tracing for agents is one of those unsexy things that quietly decides who survives production. Demos need magic. Builders need the black box to cough up receipts. x.com/JiaZhihao/stat…

@AiDevCraft @JiaZhihao It’s native support no need for a plugin.

@JiaZhihao The interesting bet here is that "open + local" outcompetes hosted observability once agents proliferate — debugging a runaway tool loop is too time-sensitive to ship traces over the wire. Does Motus capture cost/token attribution per span natively or is that a plugin layer?

🚀Introducing Motus Tracing: open-source observability for AI agents. Without traces, an agent is a black box that burns tokens. Yet most agent observability and tracing stacks today live behind accounts and subscription tiers. Motus Tracing is fully open source. Capture every model call, tool call, sandbox interaction, sub-agent action, retry, and error, for any agent framework. One unified interface from development to production. Same spans for debugging, evals, and Learning Agents. Blog: lithosai.com/blog/motus-age… Code: github.com/lithos-ai/motus @lithos_ai

@jychp_fr @JiaZhihao Please join our slack, happy to help along the way! join.slack.com/t/lithosaicomm…

@JiaZhihao Very cool, going to test this. Quick question: how do you position Motus vs DeepEval tracing or Sentry AI Observability? Feels like those stacks are starting to converge.

@samhogan @JiaZhihao Happy to chat!

@JiaZhihao this is sweet

RT @JiaZhihao: 🚀Introducing Motus Tracing: open-source observability for AI agents. Without traces, an agent is a black box that burns tok…

RT @JiaZhihao: 🚀 Introducing Learning Agents in Motus Cloud. Instead of manually tuning prompts, tools, models, and reasoning flows, Motus…

RT @JiaZhihao: 🚀Introducing Motus, the open-source agent infrastructure that learns in production. Existing agent infra serves static agen…