xerrors/Yuxi — 结合知识库管理的 多租户 Agent Harness 平台。 An agent harness that integrates a LightRAG knowledge base and knowle…

Description and use

xerrors/Yuxi is an open‑source, multi‑tenant agent harness that combines a LightRAG knowledge base with graph‑based knowledge (Neo4j) to power advanced AI agents. Built on LangChain, Vue, and FastAPI, it supports DeepAgents, MinerU PDF processing, and MCP integration, providing a unified orchestration layer for AI/ML workflows and frontend interaction.

Problem

xerrors/Yuxi helps turn isolated prompts and tools into repeatable agent workflows.

Outcome

Medium: useful for prototypes or internal workflows, with dependency and maintenance checks before production.

Workflow

• coordinate multi-agent workflows
• add tool-use pipelines
• standardize agent memory

Evidence

结合知识库管理的 多租户 Agent Harness 平台。 An agent harness that integrates a LightRAG knowledge base and knowledge graphs. Build with LangChain + Vue + FastAPI, support DeepAgents、MinerU PDF、Neo4j 、MCP.

Source