XActions

Executes browser-based Twitter/X automation via an MCP server enabling AI agents to control actions without API keys.

javascript

67

GitHub Stars

javascript

Language

4 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "nirholas-xactions": {
      "command": "node",
      "args": [
        "/path/to/xactions/src/mcp/server.js"
      ]
    }
  }
}

XActions provides an MCP server that lets AI agents drive Twitter/X automation directly through a Model Context Protocol interface. This enables programmatic, browser-based automation without exposing API keys, while keeping all automation client-side and auditable.

How to use

You use the MCP server by running an MCP client that communicates with the server to perform actions such as scraping profiles, retrieving followers, posting tweets, or liking content. The server exposes tools that a compliant AI agent can invoke to control XActions features. Set up your client to reference the MCP endpoint, then call the available tools to perform tasks end-to-end, such as discovering followers, following users, or scraping tweets. All actions are executed via the local MCP server process, ensuring your data never leaves your machine when you run with a client locally.

Available MCP tools

The MCP server provides a suite of tools your AI agent can call. Examples include: x_login to authenticate with a session, x_get_profile to fetch a user’s profile, x_get_followers and x_get_following to scrape lists, x_get_non_followers to identify non-followers, x_get_tweets to scrape a user’s tweets, x_search_tweets to query tweets, x_follow to follow a user, x_unfollow to unfollow, x_post_tweet to post a tweet, x_like to like a tweet, and x_retweet to repost.

CLI and automation flow with MCP

While you can interact with XActions through various interfaces, the MCP server enables AI agents to orchestrate actions via a standardized endpoint. Your agent can request profile data, social interactions, and content retrieval, then execute follow/unfollow and engagement actions as needed. The important pattern is that your agent issues high-level intents to the MCP tools, and the server handles the browser-driven automation locally.

Available tools

x_login

Login using a saved session cookie to authorize subsequent actions.

x_get_profile

Retrieve profile information for a given username.

x_get_followers

Scrape and return the list of followers for a user.

x_get_following

Scrape and return the list of accounts the user is following.

x_get_non_followers

Identify followers who do not follow back.

x_get_tweets

Scrape tweets from a specified user.

x_search_tweets

Search tweets by a query string.

x_follow

Follow a specified user.

x_unfollow

Unfollow a specified user.

x_post_tweet

Post a tweet on behalf of the authenticated user.

x_like

Like a specified tweet.

x_retweet

Retweet a specified tweet.