X (Twitter) Public Intel - MCP Server

Model Context Protocol (MCP) server wrapper for x-twitter-intel. Gives AI agents direct, pay-per-call access to logged-out X / Twitter profile intel, post lookups, and engagement-velocity signals - no login, no API key required, and token-less per-call payment over x402 / Skyfire.

What it is

A thin MCP server that exposes the X public-intel engine as four typed tools an AI agent can call. Each tool runs the upstream x-twitter-intel actor against X's public, logged-out syndication surface and reshapes the result into small, deterministic JSON (no megabyte payloads). Built for social-monitoring copilots, competitive-intel agents, and brand-watch bots.

Tools

Tool	What it does
get_profile_intel	Slim public profile snapshot for one handle (followers, bio, cadence) plus its engagement-velocity signal and a momentum flag.
get_recent_posts	The most recent public posts for one handle, newest first, as slim post records with full profile-timeline metrics.
get_posts	Resolve 1-50 tweet ids or status URLs to slim post records (per-post surface exposes like_count + reply_count).
compare_profiles	Compare 2-5 handles, ranked by engagement-per-follower, with a one-line comparison summary naming the leader.

Run modes

• list_tools - emit the tool catalog (free, no charge) including the agentic-payment descriptor.
• call_tool - invoke one tool. Requires tool + args.
• batch - invoke up to 10 {tool, args} calls in one run.

{
  "mode": "call_tool",
  "tool": "get_profile_intel",
  "args": { "handle": "nasa", "max_posts": 20 }
}

{
  "mode": "call_tool",
  "tool": "compare_profiles",
  "args": { "handles": ["nasa", "spacex", "esa"] }
}

Honest scope and limits

• About 20 most recent posts per profile. That is roughly what X's public, logged-out syndication surface renders. There is no public "load more" without a login, so this server does not pretend to firehose. max_posts is capped at 20. This is an intel tool, not a volume scraper.
• Per-post surface is partial. get_posts (tweet ids / status URLs) returns the metrics X's per-post surface exposes - like_count and reply_count - but not separate retweet/quote counts. Use get_recent_posts for full per-post metrics.
• Private accounts are not readable. Protected (private) accounts are not part of the public surface; they come back as an error, not data.
• Rate limits happen. The profile-timeline surface blocks bare datacenter IPs aggressively. The upstream actor handles retries and proxy rotation; a hard block becomes an error message, never a crashed run.

Responsible use / data scope

This server reads only logged-out, public X / Twitter data - the same data X serves to anonymous visitors through its embeddable widgets. No accounts, no cookies, no session tokens, no paid X API, no paywall bypass. PII is minimized to what X already shows publicly (handle, display name, public bio, public counts). This actor is not affiliated with, endorsed by, or sponsored by X Corp. You are responsible for lawful use of the outputs. See the upstream actor's README for the full data-scope note.

Agentic payments (x402 + Skyfire)

This server is x402 (USDC on Base) and Skyfire ready. When the operator enables Apify MCP monetization, an AI agent can pay per tool call with no pre-provisioned API token. The list_tools response includes a payments descriptor advertising the accepted rails and per-call price. Operators enable rails via environment variables:

Env	Purpose
X402_ENABLED	1 to advertise x402 acceptance
X402_PAY_TO_ADDRESS	receiving wallet (USDC on Base)
X402_PRICE_USD	per-call price advertised (default 0.005)
SKYFIRE_ENABLED	1 to advertise Skyfire acceptance
SKYFIRE_SELLER_ID	Skyfire seller identity

Calls through Apify always bill via standard PPE (mcp_tool_call $0.005 + upstream pass-through); x402/Skyfire are the token-less rails layered on top for direct agents.

Pricing

Flat $0.005 per MCP tool call, plus the upstream x-twitter-intel PPE pass-through (profile_record, post_record, engagement_velocity_signal) billed to the same run. list_tools is free. A run that returns nothing costs nothing beyond the tool call you triggered.

Use cases

• A brand-watch agent calls get_profile_intel on a competitor handle to read followers, cadence, and whether a post is taking off (momentum flag).
• A competitive-intel agent calls compare_profiles across a peer set to rank who gets the most engagement per follower.
• A research agent resolves a list of viral tweet URLs with get_posts to pull their text and like/reply counts.
• A monitoring copilot polls get_recent_posts to summarize what a handle has posted since the last check.

Related Actors

• x-twitter-intel - the non-MCP actor with full input controls, brand-monitor schedule mode, follower/new-post deltas, and CSV/HTML artifacts.
• reddit-topic-watcher - watch a topic across Reddit the same way.

Found this useful?

Leave a quick review: https://apify.com/seibs.co/mcp-x-twitter-intel#reviews