Skip to main content

AI MCN 自动推广工具 — 多Agent内容营销自动化,支持掘金/知乎/Dev.to

Project description

AgentCrew MCN 🤖

Your AI marketing team. Works 24/7. Never asks for a raise.

Python Version License: MIT PyPI version CI Tests Docs Wiki Coverage

📖 中文文档 | English


What is AgentCrew?

AgentCrew MCN is an open-source multi-agent content marketing automation tool. A team of AI "employees" handles the entire pipeline — from content creation to cross-platform distribution.

  • Writer Agent — Copywriter: generates technical articles, social posts, and threads
  • Reviewer Agent — QA: pre-publish content safety and quality review
  • Publisher Agent — Operations: publishes to Juejin, Zhihu, Dev.to, and more
  • Analyst Agent — Data Analyst: tracks performance and smart scheduling

Features

  • 🧠 4 AI Agents — Writer, Reviewer, Publisher, Analyst — each operates independently
  • 🔧 Skills + Tools System — Atomic operations composed into high-level capabilities (v0.4: LLM-driven)
  • 📚 RAG Knowledge Base — Retrieval-augmented generation from historical content
  • 🌍 Cross-Platform Publishing — Juejin, Zhihu, Dev.to with platform-specific style adaptation
  • 🔌 Multi-Provider LLM — DeepSeek, OpenAI, Anthropic, Ollama — switch with one config field
  • 🖼️ Image Generation — DALL-E 3 cover images for articles
  • Smart Scheduling — Analyst predicts best publish times with random jitter anti-detection
  • 📊 Dashboard + Growth Tracking — Web UI for analytics + GROWTH.md for metrics
  • 🐳 Docker — One-command deployment with docker-compose
  • 🔁 Dogfooding — The project promotes itself

Quick Start

Option 1: pip install (recommended)

# 1. Install
pip install agentcrew-mcn

# 2. Initialize config
agentcrew-mcn init

# 3. Edit .env with your API keys
#    DEEPSEEK_API_KEY=sk-...

# 4. Start using
agentcrew-mcn write generate --topic "Python async programming" --style technical
agentcrew-mcn publish post --file article.md --platform juejin --dry-run

💡 agent-crew is also available as an alias for agentcrew-mcn.

Option 2: From source

# 1. Clone
git clone https://github.com/super-rick/agentcrew-mcn.git
cd agentcrew-mcn

# 2. Install dependencies
pip install -r requirements.txt

# 3. Initialize
python -m cli.main init

# 4. Edit .env with your API keys

Configuration

agentcrew-mcn init creates a config template. Key settings:

# config.yaml
llm:
  api_key: ${DEEPSEEK_API_KEY}  # Reads from .env
  model: deepseek-chat

platforms:
  juejin:
    cookie: ${JUEJIN_COOKIE}    # Export cookie from browser
  devto:
    api_key: ${DEVTO_API_KEY}   # https://dev.to/settings/extensions

Usage

# Generate a technical article
agentcrew-mcn write generate --topic "Python async programming" --style technical

# Preview mode (no LLM call, inspect parameters)
agentcrew-mcn write generate --topic "Python async" --dry-run

# Publish to Juejin
agentcrew-mcn publish post --file article.md --platform juejin

# Scheduled publishing (every 6 hours)
agentcrew-mcn schedule start --topic-file topics.txt --platform juejin --interval 6

# RAG knowledge base
agentcrew-mcn rag ingest --file article.md --source "my_blog"
agentcrew-mcn rag search --query "AI Agent architecture"

Architecture

┌─────────────────────────────────────────────────────┐
│                   Orchestrator                       │
│           Task dispatch / Scheduler / Config          │
└──┬──────────┬──────────────┬────────────────────────┘
   │          │              │
┌──▼──────┐ ┌▼────────┐ ┌──▼────────┐
│ Writer  │ │Publisher│ │  Analyst  │
│ Agent   │ │Agent    │ │  Agent    │
│         │ │         │ │           │
│ Skills: │ │ Skills: │ │ Skills:   │
│ trending│ │ schedule│ │ report    │
│ tech    │ │ multi   │ │ recommend │
│ thread  │ │ batch   │ │           │
│         │ │         │ │           │
│ Tools:  │ │ Tools:  │ │ Tools:    │
│ search  │ │ devto   │ │ analytics │
│ rag     │ │ juejin  │ │ compare   │
│ compose │ │ zhihu   │ │ report    │
└─────────┘ └─────────┘ └───────────┘

Dashboard

AgentCrew includes a Streamlit Dashboard for monitoring your AI marketing team.

Dashboard Overview

More Dashboard Pages
Publish Analysis AI Analysis
Publishing Analytics

Tech Stack

Component Technology
Language Python 3.10+
CLI Click + Rich
LLM DeepSeek API (OpenAI-compatible)
Vector DB ChromaDB
Automation Playwright
Dashboard Streamlit + Plotly

Roadmap

See ROADMAP.md for details.

  • v0.1 ✅ Writer + Publisher + CLI + Juejin/Zhihu + Dashboard
  • v0.2 ✅ Dev.to adapter + RAG + MCP + Bilingual README
  • v0.3 ✅ Reviewer Agent + Retry + Persistence + Docker (332 tests)
  • v0.4 ✅ LLM-driven Skills + Multi-provider + Image gen + Smart scheduling
  • v0.5 ✅ 6 new platforms (CSDN/WeChat/SegmentFault/X/Xiaohongshu/Medium)
  • v1.0 🔮 Web API + Plugins + Community

License

MIT

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

agentcrew_mcn-0.5.3.tar.gz (92.8 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

agentcrew_mcn-0.5.3-py3-none-any.whl (120.0 kB view details)

Uploaded Python 3

File details

Details for the file agentcrew_mcn-0.5.3.tar.gz.

File metadata

  • Download URL: agentcrew_mcn-0.5.3.tar.gz
  • Upload date:
  • Size: 92.8 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.13.14

File hashes

Hashes for agentcrew_mcn-0.5.3.tar.gz
Algorithm Hash digest
SHA256 fed9db791ec620af36e7313d9d2eba100f63c7de8e734fccb08f0da5f051051a
MD5 32de47a698c6095b46a46395188e1791
BLAKE2b-256 8e531fb0ab390f4e1b5fdb80aef0cd7e6d28bb150af00eef2397b049bae8fb54

See more details on using hashes here.

Provenance

The following attestation bundles were made for agentcrew_mcn-0.5.3.tar.gz:

Publisher: publish.yml on super-rick/agentcrew-mcn

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file agentcrew_mcn-0.5.3-py3-none-any.whl.

File metadata

  • Download URL: agentcrew_mcn-0.5.3-py3-none-any.whl
  • Upload date:
  • Size: 120.0 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.13.14

File hashes

Hashes for agentcrew_mcn-0.5.3-py3-none-any.whl
Algorithm Hash digest
SHA256 791b406d3e74af46cf940a55c3ea6e4e57d8e973daffc647ae8b6f4f6dbc591f
MD5 e2caf5776fa33ee51e5b6fe383524d2f
BLAKE2b-256 cbdfd949648a731614293a71ed83631c0d6852f0009fff61db37efb75420ac19

See more details on using hashes here.

Provenance

The following attestation bundles were made for agentcrew_mcn-0.5.3-py3-none-any.whl:

Publisher: publish.yml on super-rick/agentcrew-mcn

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page