feedgrab-setup

Original：🇺🇸 English
Translated

Install and configure feedgrab — the universal content grabber. Use when user needs to set up feedgrab, configure API keys, login to platforms, or diagnose issues.

2installs
Sourceibigqiang/feedgrab
Added on

NPX Install

npx skill4agent add ibigqiang/feedgrab feedgrab-setup

Tags

Translated version includes tags in frontmatter
feedgrabcontent-scrapingcli-toolinstallation-guidepython-tool

SKILL.md Content

View Translation Comparison →

feedgrab Setup Guide

Install feedgrab, configure API keys, login to platforms, and verify everything works.

Trigger

  • /feedgrab-setup
  • "Install feedgrab"
  • "配置 feedgrab"
  • When 
    /feedgrab
    reports "feedgrab not found"

Step 1: Check Python Environment

bash
python3 --version 2>/dev/null || python --version 2>/dev/null
pip3 --version 2>/dev/null || pip --version 2>/dev/null
Requires Python ≥ 3.10. If not available, tell user to install Python first.

Step 2: Install feedgrab

bash
pip install feedgrab[all]
This installs the full package with all optional dependencies (browser, stealth, Twitter, WeChat, Xiaohongshu, Feishu, Telegram).
For a minimal install:
bash
pip install feedgrab                     # Core only (Jina fallback)
pip install feedgrab[browser]            # + Playwright
pip install feedgrab[twitter]            # + Twitter GraphQL
pip install feedgrab[stealth]            # + Anti-detection (patchright, curl_cffi)
After installation, verify:
bash
feedgrab --help

Step 3: Interactive Setup

bash
feedgrab setup
This runs a 5-step interactive wizard:
  1. Output directory — where to save fetched content (default: 
    ./output/
    )
  2. Jina API — free, no key needed
  3. YouTube API Key — optional, for YouTube search
  4. GitHub Token — optional, for higher rate limits
  5. Creates 
    .env
     file with your configuration

Step 4: Platform Login (as needed)

For platforms requiring cookies/session:
bash
# Twitter/X — opens browser for login
feedgrab login twitter

# Xiaohongshu — opens browser for login
feedgrab login xhs

# WeChat MP backend — opens browser for login (session valid ~4 days)
feedgrab login wechat

# Feishu — opens browser for login
feedgrab login feishu
Tip: If you have Chrome running with remote debugging enabled, set 
CHROME_CDP_LOGIN=true
in 
.env
to extract cookies from your existing browser session without re-login.

Step 5: Verify Installation

bash
# Full diagnostic
feedgrab doctor

# Platform-specific diagnostic
feedgrab doctor x          # Twitter/X
feedgrab doctor xhs        # Xiaohongshu
feedgrab doctor mpweixin   # WeChat
feedgrab doctor feishu     # Feishu

Step 6: Test

bash
# Test with a simple URL (no login needed)
feedgrab https://github.com/iBigQiang/feedgrab
If the output 
.md
file is generated successfully, setup is complete!

Advanced Configuration (.env)

Key environment variables (all optional):
env
# Output
OUTPUT_DIR=./output

# Twitter
X_BOOKMARKS_ENABLED=true
X_USER_TWEETS_ENABLED=true
X_LIST_TWEETS_ENABLED=true
X_DOWNLOAD_MEDIA=true

# Xiaohongshu
XHS_USER_NOTES_ENABLED=true
XHS_SEARCH_ENABLED=true
XHS_DOWNLOAD_MEDIA=true
XHS_FETCH_COMMENTS=true

# WeChat
MPWEIXIN_DOWNLOAD_MEDIA=true
MPWEIXIN_FETCH_COMMENTS=true

# YouTube
YOUTUBE_API_KEY=your_key_here

# GitHub
GITHUB_TOKEN=your_token_here

# Feishu
FEISHU_APP_ID=your_app_id
FEISHU_APP_SECRET=your_secret
FEISHU_DOWNLOAD_IMAGES=true

# Stealth
CHROME_CDP_LOGIN=true
CHROME_CDP_PORT=9222
See 
.env.example
in the feedgrab repo for the full list.

Troubleshooting

IssueSolution
pip install
fails		Try 
pip install --user feedgrab[all]
or use a virtualenv
Playwright not workingRun 
playwright install chromium
patchright not workingRun 
patchright install chromium
Cookie expiredRe-run 
feedgrab login <platform>
Windows encoding errorsfeedgrab auto-fixes UTF-8, but ensure Python ≥ 3.10