Facebook Scraper3 MCP Server
# Facebook Scraper3 MCP Server
[English](./README_EN.md) | 简体中文 | [繁體中文](./README_ZH-TW.md)
用于访问 Facebook Scraper3 API 的 MCP 服务器。
## 🚀 使用 EMCP 平台快速体验
**[EMCP](https://sit-emcp.kaleido.guru)** 是一个强大的 MCP 服务器管理平台,让您无需手动配置即可快速使用各种 MCP 服务器!
### 快速开始:
1. 🌐 访问 **[EMCP 平台](https://sit-emcp.kaleido.guru)**
2. 📝 注册并登录账号
3. 🎯 进入 **MCP 广场**,浏览所有可用的 MCP 服务器
4. 🔍 搜索或找到本服务器(`bach-facebook_scraper3`)
5. 🎉 点击 **"安装 MCP"** 按钮
6. ✅ 完成!即可在您的应用中使用
### EMCP 平台优势:
- ✨ **零配置**:无需手动编辑配置文件
- 🎨 **可视化管理**:图形界面轻松管理所有 MCP 服务器
- 🔐 **安全可靠**:统一管理 API 密钥和认证信息
- 🚀 **一键安装**:MCP 广场提供丰富的服务器选择
- 📊 **使用统计**:实时查看服务调用情况
立即访问 **[EMCP 平台](https://sit-emcp.kaleido.guru)** 开始您的 MCP 之旅!
---
## 简介
这是一个 MCP 服务器,用于访问 Facebook Scraper3 API。
- **PyPI 包名**: `bach-facebook_scraper3`
- **版本**: 1.0.0
- **传输协议**: stdio
## 安装
### 从 PyPI 安装:
```bash
pip install bach-facebook_scraper3
```
### 从源码安装:
```bash
pip install -e .
```
## 运行
### 方式 1: 使用 uvx(推荐,无需安装)
```bash
# 运行(uvx 会自动安装并运行)
uvx --from bach-facebook_scraper3 bach_facebook_scraper3
# 或指定版本
uvx --from bach-facebook_scraper3@latest bach_facebook_scraper3
```
### 方式 2: 直接运行(开发模式)
```bash
python server.py
```
### 方式 3: 安装后作为命令运行
```bash
# 安装
pip install bach-facebook_scraper3
# 运行(命令名使用下划线)
bach_facebook_scraper3
```
## 配置
### API 认证
此 API 需要认证。请设置环境变量:
```bash
export API_KEY="your_api_key_here"
```
### 环境变量
| 变量名 | 说明 | 必需 |
|--------|------|------|
| `API_KEY` | API 密钥 | 是 |
| `PORT` | 不适用 | 否 |
| `HOST` | 不适用 | 否 |
### 在 Cursor 中使用
编辑 Cursor MCP 配置文件 `~/.cursor/mcp.json`:
```json
{
"mcpServers": {
"bach-facebook_scraper3": {
"command": "uvx",
"args": ["--from", "bach-facebook_scraper3", "bach_facebook_scraper3"],
"env": {
"API_KEY": "your_api_key_here"
}
}
}
}
```
### 在 Claude Desktop 中使用
编辑 Claude Desktop 配置文件 `claude_desktop_config.json`:
```json
{
"mcpServers": {
"bach-facebook_scraper3": {
"command": "uvx",
"args": ["--from", "bach-facebook_scraper3", "bach_facebook_scraper3"],
"env": {
"API_KEY": "your_api_key_here"
}
}
}
}
```
## 可用工具
此服务器提供以下工具:
### `profiles_photos`
Read profiles photos (from recent) Use cursor to get next page of results
**端点**: `GET /profile/photos`
---
### `comment_by_id`
Get comment by comment id
**端点**: `GET /comment`
---
### `search_people`
Searches people
**端点**: `GET /search/people`
---
### `search_hashtag`
Search posts with hashtag
**端点**: `GET /search/hashtags`
---
### `profile_reels`
Get profile reels. Get `reels_profile_id` from profile details endpoint
**端点**: `GET /profile/reels`
---
### `page_events`
Get all future events created by page
**端点**: `GET /page/events/future`
---
### `post_reactions`
Get post reactions
**端点**: `GET /post/reactions`
---
### `comments___nested`
Get nested comments. You need to use legacy fb id
**端点**: `GET /post/comments_nested`
---
### `page_reels`
Get page reels by page id - use reels_page_id now
**端点**: `GET /page/reels`
---
### `listing_details`
Get listing details
**端点**: `GET /marketplace/details`
---
### `search`
Search marketplace. Use cursor to get next page
**端点**: `GET /marketplace/search`
---
### `comments`
Get post comments. Use comment_id from post
**端点**: `GET /post/comments`
---
### `search_groups_posts`
Searches group for posts
**端点**: `GET /search/groups_posts`
---
### `page_videos`
Get page videos by page delegate id. YOU NEED TO USE DELEGATE ID, NOT PAGE ID.
**端点**: `GET /page/videos`
---
### `pages_photos`
Read page photos (from recent) Use cursor to get next page of results
**端点**: `GET /page/photos`
---
### `pages_posts`
Read page posts (from recent)
**端点**: `GET /page/posts`
---
### `page_details`
Read page details
**端点**: `GET /page/details`
---
### `page_past_events`
Get all past events created by page. Use cursor to get next page of results.
**端点**: `GET /page/events/past`
---
### `get_group_posts`
Get group posts. Only PUBLIC groups can be scrapped! If there are no post response, check if group is not private.
**端点**: `GET /group/posts`
---
### `get_group_future_events`
Get group future events. Only PUBLIC groups can be scrapped! If there are no post response, check if group is not private. Use cursor to get next page of events!
**端点**: `GET /group/future_events`
---
### `search_events`
Searches for facebook events
**端点**: `GET /search/events`
---
### `game_lives`
Get game live videos Use cursor to get next page.
**端点**: `GET /gaming/live`
---
### `browse_games`
Browse games. Use empty query for browse all games. Use cursor to get next page.
**端点**: `GET /gaming/games`
---
### `page_reviews`
Get page reviews. Use cursor to get next page
**端点**: `GET /page/reviews`
---
### `search_video`
Performs facebook videos search
**端点**: `GET /search/videos`
---
### `profile_posts`
Read profile posts (from recent), if public
**端点**: `GET /profile/posts`
---
### `events_details_by_id`
Get events details by event id
**端点**: `GET /event/details_id`
---
### `locations`
Search for facebook locations id. If you get strange results, try add country to query and/or try without diacritic
**端点**: `GET /search/locations`
---
### `search_post`
Performs facebook posts search
**端点**: `GET /search/posts`
---
### `events_details_by_url`
Get events details
**端点**: `GET /event/details`
---
### `profiles_details_by_id`
Get profiles details by id
**端点**: `GET /profile/details_id`
---
### `profile_details_by_url`
Get profile details by url
**端点**: `GET /profile/details_url`
---
### `profile_id`
Get profile id by url
**端点**: `GET /profile/profile_id`
---
### `get_post_details`
Get post details by url or post id. If both are set, post id is used.
**端点**: `GET /post`
---
### `get_group_details`
Get group summary
**端点**: `GET /group/details`
---
### `get_group_id`
Getting group facebook id
**端点**: `GET /group/id`
---
### `page_id`
get page id from url
**端点**: `GET /page/page_id`
---
### `reshares`
Get post reshares
**端点**: `GET /post/share`
---
### `search_place`
Search for fb place
**端点**: `GET /search/places`
---
### `search_pages`
Searches for facebook pages
**端点**: `GET /search/pages`
---
## 技术栈
- **传输协议**: stdio
- **HTTP 客户端**: httpx
## 许可证
MIT License - 详见 [LICENSE](./LICENSE) 文件。
## 开发
此服务器由 [API-to-MCP](https://github.com/BACH-AI-Tools/api-to-mcp) 工具生成。
版本: 1.0.0
TDQS
Scored across 40 tools
The tool set has clear distinctions for major categories like pages, groups, profiles, and search, but there is significant overlap within categories that could cause confusion. For example, 'comments' and 'comments_2' have ambiguous boundaries, and multiple search tools (e.g., 'search_post', 'search_video', 'search_hashtag') might be misselected due to similar purposes. Descriptions help clarify some differences, but the overlap is notable.
Most tools follow a consistent verb_noun or noun_verb pattern with underscores, such as 'browse_games', 'get_group_details', and 'search_events'. There are minor deviations like 'page_id' (noun_noun) and 'reshares' (noun only), but overall the naming is predictable and readable across the set.
With 40 tools, the count is excessive for a scraper server, leading to potential bloat and complexity. While Facebook has many data types, this many tools suggests over-fragmentation rather than a well-scoped set, making it harder for agents to navigate efficiently. A more consolidated approach with 15-25 tools would be more appropriate.
The tool surface covers a wide range of Facebook entities comprehensively, including pages, groups, profiles, events, posts, comments, and various search functions. Minor gaps exist, such as limited update or delete operations (expected for scraping), but agents can perform most data retrieval tasks without significant dead ends in the core workflows.