automation

deer-flow: Automate Tasks

Name: deer-flow: Automate Tasks
Author: bytedance

Founders automate tasks with deer-flow, a Python SuperAgent harness. Simplify workflows with this open-source tool.

83,603 stars+264★ this week11,622 forksPythonGuide quality 8/10 · reviewed Jul 3Updated 10/10/2026100% free · open source

What it does

Deer-flow automates tasks by harnessing the power of Python SuperAgent, simplifying workflows for founders and developers.

When to use it

•When automating complex workflows with multiple dependencies
•As a replacement for manual, repetitive tasks in development and testing
•To streamline data processing and integration across different systems

Ready-to-paste prompt

deer-flow run --config deer.yaml --task my_task --param foo=bar

Heads up: Ensure Python 3.7 or later is installed, as deer-flow relies on modern Python features and may not be compatible with earlier versions

Saves to your device

Use with Claude

New

Skip the builder — one click puts this in Claude, Cursor, Antigravity and more.

🔑 Needs an API key

Works after you add credentials — the setup agent will ask you for them.

Try it instantly — no install

Claude Code

mkdir -p ~/.claude/skills/deer-flow && curl -fsSL https://workflowstacks.com/api/skills/deer-flow/claude-skill -o ~/.claude/skills/deer-flow/SKILL.md

Open in another AI app

Opens the app with this repo with the prompt ready to go — no copy-paste needed.

Connect the whole catalog (MCP)

claude mcp add --transport http workflowstacks https://workflowstacks.com/api/mcp

Adds a WorkflowStacks connector to Claude Code: search and load any skill here by chatting.

How deer-flow: Automate Tasks works

Codeflow

Free to inspect

Deer-flow: Automate Tasks is a very large Python project (~226k lines across 1090 code files, plus 733 test files). Setup is light: installs like a normal app — add your API keys. Reading the code is optional. Last commit this month, MIT license, has a test suite.

Size

Very large codebase

~226k lines · 1090 code files · days to read — use, don't read

Setup

One-command install

Installs like a normal app — add your API keys. Reading the code is optional.

Runs on

Docker

Needs API keys (.env)

Python 82%TypeScript 14%MDX 2%HTML 1%Shell 1%

Where to start reading

1
README.md
Start here — what it does and how to install it
2
AGENTS.md
The instructions the AI actually follows
3
.env.example
The API keys and settings you must provide
4
skills/public/academic-paper-review/SKILL.md
Inside skills/ — an example of what the AI is told to do
5
examples/deerflow-extension-example/README.md
A worked example — copy this to get going

What's in each folder

skills/Prompts, skills & agent definitions104 files

backend/Backend / API1250 files

frontend/Frontend / UI738 files

docs/Documentation41 files

examples/Examples you can copy6 files

scripts/Helper scripts43 files

deploy/Deployment / infrastructure27 files

docker/Deployment / infrastructure19 files

READMEHas testsDocumentedCI checksDocker readyExamples includedMIT licenseUpdated this month

Use with Claude

New

Skip the builder — one click puts this in Claude, Cursor, Antigravity and more.

🔑 Needs an API key

Works after you add credentials — the setup agent will ask you for them.

Try it instantly — no install

Claude Code

mkdir -p ~/.claude/skills/deer-flow && curl -fsSL https://workflowstacks.com/api/skills/deer-flow/claude-skill -o ~/.claude/skills/deer-flow/SKILL.md

Open in another AI app

Opens the app with this repo with the prompt ready to go — no copy-paste needed.

Connect the whole catalog (MCP)

claude mcp add --transport http workflowstacks https://workflowstacks.com/api/mcp

Adds a WorkflowStacks connector to Claude Code: search and load any skill here by chatting.

Quick Actions

Details

Creator

bytedance

Language

Python

Do this with a working workflow

Pages we wrote and tested that put a skill like this to work — importable n8n templates, outcome guides and Claude Desktop configs.

Outcome guide

Automate your meeting summaries

A private form takes any transcript (from Zoom, Meet, Otter, Fireflies) and emails you back a structured summary: a TL;DR, the decisions made, and action items with owners.

Outcome guide

Turn meeting notes into a summary email automatically

Paste the transcript into a private form and the workflow writes and emails a concise recap — decisions, owners, open questions — that people will actually read.