Overview
[Product] DeerFlow 2.0 is an open-source super Agent harness that gives Agents the full infrastructure needed for complex work. Rebuilt on LangGraph and LangChain, it grew from a single deep-research tool into a full-stack runtime. DeerFlow coordinates sub-Agents, memory modules, and secure sandboxes so Agents can plan, break down tasks, and actually execute—covering automated workflows from data analysis to content creation. [Features] Extensible skills: define Agent Skills (search, report generation, and more) in Markdown with on-demand loading and MCP server support. Sub-Agent orchestration: the main Agent can dynamically split hard work, spawn sub-Agents in isolated contexts in parallel, and merge results. Full sandbox: a Docker-isolated environment with its own filesystem, code execution, and CLI. Memory and context: long-term memory across sessions, plus context summaries that keep long-running jobs efficient. Open-source: https://github.com/bytedance/deer-flow
Connected profiles
More Coding assistants products
Sources · 0Sources & provenance
English text is maintained separately from the original record. Consult the original source before citing a fact.
No source links are available for this profile.
Original source text
Names may use phonetic renderings when no English name is supplied. The original wording is retained below.
【产品介绍】DeerFlow 2.0 是一个开源的超级 Agent 治理框架（Harness），旨在为 Agent 提供执行复杂任务所需的完整基础设施。它基于 LangGraph 和 LangChain 重构，从单一的深度研究工具进化为全栈运行时环境。DeerFlow 能够协调子 Agent、记忆模块和安全沙箱，使 Agent 具备规划、拆解任务及实际执行工作的能力，适用于从数据分析到内容创作的各类自动化工作流。【产品功能】可扩展技能系统：支持通过 Markdown 定义 Agent Skill（如搜索、报告生成），具备按需加载机制及 MCP 服务器支持。子 Agent 编排：主 Agent 可动态将复杂任务拆解，生成子 Agent 在独立上下文中并行处理并汇总结果。全功能沙箱：提供基于 Docker 的隔离环境，赋予 Agent 独立的文件系统、代码执行及命令行操作能力。记忆与上下文管理：具备跨会话的长期记忆能力，并通过上下文摘要优化长流程任务的执行效率。项目开源地址：https://github.com/bytedance/deer-flow