All activity
Treadstone is an agent-native sandbox platform for longer-running AI work. You can run agents inside isolated sandboxes, or treat sandboxes as tools that agents call on demand. Each sandbox is built around an all-in-one runtime: code execution, shell, file system, browser-facing surfaces, MCP, and long-running state.
Treadstone is open source, self-hostable, and built on Kubernetes.
TreadstoneAI Agent Sandbox
Bryce Yuleft a comment
I believe that 2026 will be the year of the AI agent. I have always wanted to build my own multi-AI agent system, which led me to develop my own AI agent infrastructure. Eventually, I thought: why not turn this into a product so that everyone can use it with me? If you are interested in the AI sandbox and believe in the two paths of "Agent in Sandbox" and "Sandbox as a Tool," then please be...
TreadstoneAI Agent Sandbox
Bryce Yuleft a comment
ApeRAG is a fully open-source, easy-to-deploy Agentic Graph RAG platform. We started as LightRAG users, but quickly ran into limitations: no knowledge base isolation, no concurrency (so document processing was painfully slow), no multi-process deployment, and no agent support. That’s why we built ApeRAG — to solve these challenges. With ApeRAG, documents are precisely parsed using MinerU and...
ApeRAGProduction-ready Graph RAG with advanced AI agents