TL;DR
X now offers a hosted MCP server that lets AI tools like Claude and Cursor connect to its API directly, joining GitHub, Slack, and Stripe in the MCP wave.
The hosted server removes the infrastructure work developers previously needed to connect AI assistants to X's real-time data
X now offers a hosted MCP server that lets AI tools like Claude and Cursor connect to its API directly, joining GitHub, Slack, and Stripe in the MCP wave.
X has launched a hosted Model Context Protocol server that lets AI tools like Claude, Cursor, and Grok Build connect directly to its API using a user’s own account permissions. The move eliminates the integration work developers previously had to do on their own, from building a custom MCP server to handling authentication. X joins a growing list of platforms including GitHub, Slack, Notion, Stripe, and Salesforce that now offer official MCP endpoints.
MCP is the open standard introduced by Anthropic in late 2024 that defines how AI models connect to external tools and services. Before X’s hosted server, developers who wanted an AI assistant to access the platform had to build their own MCP server, host it, wire it to the X API, and manage the authentication flow. That infrastructure overhead is now handled by X itself.
The hosted server does not add new API capabilities. Developers could already search posts, read timelines, look up users, and analyze conversations through the existing API. What changes is how those capabilities are exposed to AI applications.
X is actually rolling out two MCP servers: one for core API access and another that provides AI tools with direct access to X’s developer documentation. The second server lets AI tools reference API specs and integration guides programmatically during development workflows.
The spam question is the obvious concern. Removing an infrastructure barrier makes it easier for bad actors to automate posting at scale. X says its API rules still apply and that the hosted MCP does not bypass existing restrictions on spammy behavior.
The company also updated its API v2 earlier this year specifically to address AI-generated spam, particularly automated replies to conversations. It has also made spam more expensive.
In April, X raised the cost of publishing a post through the API to one and a half cents and the cost of posting a link to 20 cents, up from one cent for both. The price increases were designed to “curb vectors of misuse,” the company said. Whether pricing alone can deter determined spammers while the platform simultaneously makes AI integration frictionless is an open question.
The broader pattern is clear. Platform after platform is shipping official MCP servers, turning what was a developer side project into supported infrastructure. For X, the bet is that positioning itself as a real-time data source for AI agents is more valuable than the spam risk that comes with making that data easier to access.
Get the most important tech news in your inbox each week.