A
licenseNot graded
qualityD
maintenanceAn MCP server for interacting with the Twitter/X API v2, enabling AI assistants to retrieve tweets, post content, reply, quote, and more programmatically.
2,203 npm
13
MIT
bnbot-mcp-server
by jackleeio
Provides tools to fetch TikTok metadata and download video locally.
Provides tools to fetch and extract WeChat article content.
Provides tools to fetch Xiaohongshu note content and metadata.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@bnbot-mcp-server Scrape my Twitter timeline"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
bnbot-mcp-server
MCP Server for BNBOT Chrome Extension - Control Twitter/X via AI assistants like OpenClaw or Claude Desktop.
Architecture
AI Assistant (OpenClaw / Claude Desktop)
↓ stdio (MCP protocol)
bnbot-mcp-server (local Node.js process)
↓↑ WebSocket (ws://localhost:18900)
BNBOT Chrome Extension
↓ DOM operations
Twitter/XRelated MCP server: Twitter MCP
Setup
1. Install BNBOT Chrome Extension
Install from the Chrome Web Store.
2. Enable OpenClaw Integration
In the BNBOT extension sidebar, open Settings and turn on the OpenClaw toggle.
3. Configure Your AI Assistant
OpenClaw / Claude Desktop
Add to your MCP config:
{
"mcpServers": {
"bnbot": {
"command": "npx",
"args": ["bnbot-mcp-server"]
}
}
}Custom Port
{
"mcpServers": {
"bnbot": {
"command": "npx",
"args": ["bnbot-mcp-server", "--port", "9999"]
}
}
}4. Use It
Ask your AI assistant:
"Scrape my Twitter timeline"
"Search for tweets about AI"
"Post a tweet saying hello world"
"Navigate to my bookmarks and scrape them"
Available Tools
Scrape
Tool | Description |
| Scrape tweets from the timeline |
| Scrape bookmarked tweets |
| Search and scrape results |
| Scrape currently visible tweets |
| Scrape a tweet thread from URL |
| Get account analytics data |
Tweet
Tool | Description |
| Post a tweet with optional images |
| Post a thread of tweets |
| Reply to a tweet |
| Quote a tweet with optional comment |
Engagement
Tool | Description |
| Like a target tweet |
| Retweet a target tweet |
| Follow a target user |
Navigation
Tool | Description |
| Go to a specific tweet |
| Go to search page |
| Go to bookmarks |
| Go to notifications |
| Go to following timeline |
| Go back to timeline |
Status
Tool | Description |
| Check extension connection |
| Get current page info |
Content
Tool | Description |
| Fetch and extract WeChat article content |
| Fetch TikTok metadata and download video locally |
| Fetch Xiaohongshu note content and metadata |
Article
Tool | Description |
| Open X article editor via stable flow |
| Fill article title field |
| Fill article body ( |
| Upload article header image |
| Publish article (or keep draft) |
| End-to-end article creation flow |
Testing
Use the MCP Inspector to test tools directly:
npx @modelcontextprotocol/inspector npx bnbot-mcp-serverLicense
MIT
Available Tools
31 toolsaccount_analyticsC
Get Twitter/X account analytics data for a date range.
Inputs: endDate, startDate, granularity.
create_articleA
Create and publish a Twitter/X article (long-form content). Handles the full flow: open editor, fill title, fill body, and optionally publish.
Inputs: title (required), content (required), format, publish, +2 more.
fetch_tiktok_videoB
Fetch TikTok video metadata including description, author info, and video download URL. Automatically downloads the video via the browser extension and saves it locally.
Inputs: url (required), savePath.
fetch_wechat_articleB
Fetch and extract content from a WeChat Official Account article. Returns markdown-formatted article content including title, author, and body text.
Inputs: url (required).
fetch_xiaohongshu_noteA
Fetch content from a Xiaohongshu (Little Red Book) note including text, images, author, and engagement metrics.
Inputs: url (required).
fill_article_bodyB
Fill article body in the current editor. Supports optional body images.
Inputs: content (required), format, bodyImages.
fill_article_titleB
Fill article title in the current Twitter/X article editor.
Inputs: title (required).
follow_userA
Follow a user on Twitter/X. If username is provided, navigates to their profile first. Otherwise follows the author of the currently open tweet.
Inputs: username.
get_current_page_infoB
Get information about the current page open in the Twitter/X tab (URL, page type, etc.).
Inputs: no named parameters.
get_extension_statusB
Check if the BNBOT Chrome extension is connected and get its current status.
Inputs: no named parameters.
like_tweetA
Like the tweet on the currently open tweet page. Navigate to the tweet first using navigate_to_tweet.
Inputs: no named parameters.
open_article_editorA
Open the Twitter/X article editor using the stable flow (navigate to compose page then click create).
Inputs: no named parameters.
post_threadC
Post a thread (multiple connected tweets) on Twitter/X.
Inputs: tweets (required), draftOnly.
post_tweetC
Post a new tweet on Twitter/X. Supports text and optional images/videos.
Inputs: text (required), media, draftOnly.
publish_articleC
Publish article or keep as draft (auto-save).
Inputs: asDraft, publish.
quote_tweetA
Quote tweet the currently open tweet with custom text. Navigate to the tweet first using navigate_to_tweet.
Inputs: text (required), media, draftOnly.
return_to_timelineA
Navigate back to the Twitter/X home timeline.
Inputs: no named parameters.
retweetA
Retweet/repost the tweet on the currently open tweet page. Navigate to the tweet first using navigate_to_tweet.
Inputs: no named parameters.
scrape_bookmarksA
Scrape bookmarked tweets from Twitter/X. Automatically navigates to the bookmarks page.
Inputs: limit.
scrape_current_viewA
Scrape tweets currently visible on the page without scrolling.
Inputs: no named parameters.
scrape_search_resultsB
Search for tweets and scrape the results. Navigates to search page with the given query.
Inputs: query (required), limit.
scrape_threadA
Scrape all tweets from a Twitter/X thread. Must be on a tweet detail page (/status/ URL). Collects all tweets by the same author, returns individual tweets and merged text.
Inputs: maxScrolls.
scrape_timelineA
Scrape tweets from the Twitter/X timeline. Returns tweet text, author, metrics, and URLs. Navigate to the timeline first if not already there.
Inputs: limit, scrollAttempts.
search_jobsB
Search for available jobs — tweet engagement campaigns with crypto rewards. Returns a list of active jobs you can complete to earn rewards.
Inputs: sort, type, limit, token, +3 more.
submit_replyB
Reply to a specific tweet. Navigate to the tweet first or provide the tweet URL.
Inputs: text (required), image, tweetUrl.
upload_article_header_imageA
Upload a header image in the current article editor. Supports URL/local path/data URL.
Inputs: headerImage (required).
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
B3.3/5.0
Scored across 31 tools
Disambiguation4/5
Each tool largely targets a distinct action or resource (navigation, scraping, posting, engagement, article creation), and descriptions clarify boundaries. The only notable overlap is create_article duplicating the granular open_article_editor, fill_article_title, fill_article_body, upload_article_header_image, and publish_article tools, which could confuse an agent about which path to use.
Naming Consistency4/5
Predominantly consistent snake_case verb_noun pattern (e.g., post_tweet, scrape_timeline, navigate_to_tweet, fetch_wechat_article). Minor deviations exist: retweet lacks a noun, account_analytics has no verb, and return_to_timeline uses 'return' instead of 'navigate'. Still highly readable and mostly predictable.
Tool Count2/5
31 tools is above the typical 3-15 range, and the article workflow is unnecessarily split into six granular tools plus a redundant create_article wrapper. While the breadth of Twitter automation and cross-platform fetching justifies many tools, the count feels heavy and includes clear redundancy.
Completeness4/5
Core Twitter/X lifecycle is well covered: navigation, scraping, posting, replying, liking, retweeting, quoting, following, analytics, and long-form articles. Minor gaps exist (e.g., delete tweet, bookmark/unbookmark, direct messages, reading notifications), but agents can accomplish most stated goals.
Maintenance
ActivityInactive
ResponsivenessNo recent community issues
Related MCP Connectors
Hosted X (Twitter) MCP server for Claude, ChatGPT and Cursor, with each action switched per account.
Hosted MCP server for AI agents to post to LinkedIn profiles and X and upload to YouTube.
Hosted Web Crawler MCP server for Claude, ChatGPT and Cursor, with each action switched per account.
PostHero MCP server — schedule and publish social media posts from any MCP-compatible AI tool
Related MCP Servers
- AlicenseAqualityCmaintenanceMCP server for Twitter/X enabling AI agents to search, post, reply, and engage with tweets.1415 npm2MIT
- AlicenseBqualityDmaintenanceMCP server for interacting with the X platform (Twitter) via MCP clients like Claude, Cursor AI, and Windsurf AI.2018 npm6MIT
- AlicenseAqualityBmaintenanceMCP server to read X (Twitter) posts, threads, replies, quotes, and search using your own logged-in session, no API key required.8177 PyPI4MIT