DeerFlow is unclaimed —

DeerFlow

An open-source SuperAgent harness that researches, codes, and creates.·deerflow.tech

DeerFlow is an open-source, LangChain-based framework for building and operating agent systems ("super agents"). It provides a runtime harness with memory, tools, skills, sandboxes, and subagents for composing agents, plus a reference application layer for deployment and end-user workflows. It can execute commands, manage files, and run long tasks inside a secure Docker-based sandbox, and supports multiple LLM providers.

What it's for

Deep research report generation on a given topicGenerating a video and reference image based on a novel's scenesGenerating a comic strip explaining a technical conceptExploratory data analysis of a dataset with visualizationsWatching a video and conducting deep research on its contentCollecting and summarizing a person's public appearances into a report

Features 14

Agent Skills with progressive loading
All-in-One (AIO) Sandbox combining Browser, Shell, File, MCP, and VSCode Server
Built-in and custom skill files (SKILL.md)
Deep Research capability
Docker-based secure sandbox execution environment
Extensible skills and tools (plug/swap built-in tools)
Flexible multi-model support
Long task running with planning and sub-tasking
Long/Short-term memory (Context Engineering)
Persistent, mountable file system
Reference application layer for deployment and operations
Self-hosted deployment with full control
Sequential or parallel task execution
Subagents support

At a glance

free tieropen sourceself-hostableAPI

TypeLibrary / SDK

DeploymentSelf-hosted

Fordevelopers, teams building custom agent products

Integrations

LangChainDoubaoDeepSeekOpenAIGeminiDockerVSCode ServerMCP

Resources

Docs

Pricing

Product is presented as free and open source (MIT License); no paid plans or pricing tiers are mentioned.

Open SourceFree

Described on the homepage as 'Free Open Source, MIT License, self-hosted, full control'.