bluex-mcp
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., "@bluex-mcp search X for latest on @karpathy last 7 days"
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.
bluex-mcp
Live X (Twitter) + web search MCP server β powered by your locally logged-in Grok CLI, running on your X Premium / SuperGrok subscription quota.
Your coding agent doesn't know what shipped this morning. It doesn't see the X thread where ten other developers are hitting the same bug. It can't watch the keynote that started 20 minutes ago, can't follow the leak about your competitor's pricing, can't tell you what @karpathy posted this week.
bluex-mcp closes that gap. If you pay for X Premium ($8/mo), X Premium+ ($40/mo), or SuperGrok ($30/mo), you already have first-class live X + web access through Grok. This is a thin MCP server that exposes it to Claude Code, Codex, Cursor, Claude Desktop, and any other MCP client β by driving your locally logged-in grok CLI. Usage counts against your existing plan.
Why
Other Grok MCP servers use the xAI API. You need a key, a funded account, and you pay per token. bluex-mcp takes a different path.
xAI API (most Grok MCPs) | bluex-mcp | |
Auth |
| Your logged-in |
Cost | Pay-per-token API billing | $0 β runs on your subscription's quota |
Live X access | Yes | Yes (same Grok) |
Setup | Get a key, fund a balance |
|
Footprint | β | 2 search tools + 1 health check |
If you don't already pay for Grok, an API-based MCP is the right call. If you do,
bluex-mcpmakes that subscription do double duty.
Related MCP server: grok-search-mcp
Quick start
npm install -g github:Derek-Wong213/bluex-mcp
grok login
bluex-mcp healthWire it into your agent:
# Claude Code
claude mcp add bluex -- bluex-mcp# Codex (~/.codex/config.toml)
[mcp_servers.bluex]
command = "bluex-mcp"// Cursor / Claude Desktop (mcp.json)
{ "mcpServers": { "bluex": { "command": "bluex-mcp" } } }Restart. Ask "search X for what people are saying about <topic> in the last hour" β it routes through bluex_x_search.
What people actually ask it
Each example is a real moment where an agent without live X access just stalls. They all route through the two search tools below.
Breaking AI / model release, first signal.
"What did @AnthropicAI just announce? Last hour."
The newest model dropped on X 30 minutes ago; your agent's knowledge cutoff is six weeks behind. β bluex_x_search Β· recency=1hr Β· from:@handle
A library or CLI just broke and the docs are clueless.
"X mentions of cursor composer-2.5-fast billing in the last 4 hours?"
The "me too" wave shows up on X a full day before GitHub Issues triages it. β bluex_x_search recency=4hr β bluex_web_search the issue tracker
Vendor support went silent.
"Timeline + verbatim user quotes on the Cursor on-demand billing complaints this week."
The refund ticket disappeared into a black hole; the real signal lives in X user threads. β bluex_x_search mode=deep returns timeline + verbatim quotes + sentiment
Track a thinker, not a feed.
"From @karpathy, last 7 days, anything on nano-banana."
A specific account speaking up is what unblocks the call. β bluex_x_search from:@handle last:7d
Conference happening right now.
"#AnthropicDevDay tweets from the last hour."
The keynote is live; the official blog will recap six hours later. β bluex_x_search with hashtag + last:1hr
Competitor pre-launch intel.
"What's on X about Cursor pricing changes this week?"
Roadmap leaks land on X before any press release. β bluex_x_search recency=7d mode=deep
Breaking change before the docs admit it.
"Anyone hitting issues with Next.js 16 app router migration in the last day?"
The library's docs say "stable"; X has the unfixed-bug thread. β bluex_x_search + bluex_web_search on the upstream issue
Indie launch positioning research.
"Recent MCP server launch posts on X, last 14 days β what got real engagement?"
Studying how others framed a similar product before you ship yours. β bluex_x_search with engagement cues in the prompt
Academic paper reaction loop.
"Show me the X discussion around the <paper title> arXiv preprint."
Authors don't blog; the actual debate plays out in quote-tweets. β bluex_x_search mode=deep for perspectives + verification
Tools
Tool | Purpose |
| Search X/Twitter β filter by |
| Search the live web β domain include/exclude, recency |
| Check grok install, model, profile, login status |
Search modes
Mode | What you get |
| Summary + source items (title, url, author, handle, timestamp, snippet) + citations |
| Basic + |
Example output
bluex_x_search Β· mode: "basic"
{
"summary": "β¦",
"items": [
{ "title": "β¦", "url": "https://x.com/β¦/status/β¦", "handle": "@β¦",
"publishedAt": "2026-06-30T12:00:00Z", "snippet": "β¦" }
],
"citations": [{ "url": "https://x.com/β¦", "domain": "x.com", "secure": true }],
"urls": ["https://x.com/β¦"],
"diagnostics": {
"mode": "x", "analysisMode": "basic", "parseOk": true, "structured": true,
"model": "grok-build", "profile": "shared", "attempts": 1, "warnings": []
}
}mode: "deep" adds:
{
"analysis": {
"timeline": [{ "date": "β¦", "event": "β¦", "source": "β¦" }],
"quotes": [{ "quote": "β¦", "speaker": "β¦", "source": "β¦" }],
"perspectives": [{ "viewpoint": "β¦", "summary": "β¦", "sources": ["β¦"] }],
"sentiment": "β¦",
"verification": {
"confirmed": ["β¦"], "unconfirmed": ["β¦"], "contradictory": ["β¦"]
}
}
}How it works
your agent ββMCP/stdioβββΆ bluex-mcp ββchild processβββΆ grok CLI (your login) βββΆ live X + webEach search runs grok --single β¦ --output-format json --permission-mode dontAsk --json-schema β¦ in a throwaway working directory. The schema constrains the model to emit conforming JSON β no regex-guessing from prose.
Auth cache β login checked every 5 min; a cold probe takes ~2 s, then instant.
Recursion guard β
BLUEX_MCP_ACTIVEenv var prevents circular launches.Isolation β memory / subagents / plan-mode disabled; your grok sessions are untouched.
Configuration
Set these as environment variables in your MCP client's server config. All optional.
Variable | Default | Purpose |
|
| Path to the grok CLI binary |
|
| Per-call timeout (ms) |
|
| Max agentic turns the child grok may take |
|
| Retries on empty/failed runs (exponential backoff) |
| (grok default) | Reasoning effort β |
|
| Run grok under a dedicated clean profile (see below) |
|
| HOME for the isolated profile |
| (off) | Path to JSONL log file |
|
| How the |
|
| Also record per-tool dispatch counts in each log line |
|
| Master switch for the internal A/B experiment sampler (off in production) |
|
| Comma-separated effort levels the A/B sampler picks from. |
|
| Probability of injecting the X search strategy prompt when A/B is on |
Model is pinned to
grok-buildand is intentionally not configurable. The only other locally available model,grok-composer-2.5-fast, is a coding model β not suited for search.
Isolated profile (optional)
By default bluex-mcp reuses your existing Grok login. For full isolation (no grok config, rules, or other MCP servers loaded), set BLUEX_ISOLATED=1 and log in once:
BLUEX_ISOLATED=1 bluex-mcp login
# Then add BLUEX_ISOLATED=1 to your MCP client's environment for this server.FAQ
How is this different from other Grok or Twitter MCP servers? Other Grok-based MCPs are built on the xAI API β they take an XAI_API_KEY and bill per token. bluex-mcp takes a different path: it drives your already-logged-in grok CLI, so calls run on your X Premium / SuperGrok subscription quota. See the Why table for the full comparison.
Do I need an xAI API key? No. grok login with your existing subscription is all you need.
Is there a usage limit? Yes β you're subject to your Grok plan's rate limits. SuperGrok allows ~100 prompts/2 hr; lower tiers less. source This is your normal Grok quota, not an extra limit.
Does SuperGrok include an X blue check? No. SuperGrok is a standalone AI subscription with no X social features. The blue checkmark comes with X Premium / X Premium+. source
Is this affiliated with xAI or X? No. Independent open-source wrapper around the public Grok CLI.
Does it store my credentials? No. Auth lives in the grok CLI's own profile. bluex-mcp only executes the binary.
Will it interfere with my normal grok sessions? No. Each search runs in a throwaway directory with memory/subagents/plan-mode off. For full config isolation, set BLUEX_ISOLATED=1.
Acknowledgements
bluex-mcp's initial architecture was forked from z2z23n0/grok-search-mcp. The deep analysis structure draws on stat-guy/grok-search-mcp, recency presets from mzkrasner/grok-x-insights, and tool categorisation from guzus/grok-mcp.
License
MIT β LICENSE
Available Tools
3 toolsbluex_health_checkCheck bluex-mcp + grok wiringA
Verify the grok CLI is installed, which model is used, the isolated HOME, and whether you are logged in. Run this first if searches fail.
Inputs: no named parameters.
bluex_web_searchSearch the live webA
Search the live web using your logged-in Grok account. Returns a summary, source items, and citations. Set mode="deep" for timeline, quotes, perspectives, sentiment, and verification.
Inputs: query (required), mode, domains, recency, +2 more.
bluex_x_searchSearch X (Twitter) liveB
Search live X/Twitter posts and discussion using your logged-in Grok account. Returns a summary, source items (title, URL, handle, timestamp, snippet), and citations. Set mode="deep" for timeline, verbatim quotes, multiple perspectives, sentiment, and verification.
Inputs: query (required), mode, since, until, +3 more.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelogβ¦
TDQS
Scored across 3 tools
Each tool has a unique and non-overlapping purpose: health check for system readiness, web search for general web queries, and X search for Twitter-specific content. There is no ambiguity in their intended use.
All tool names follow a consistent pattern with the 'bluex_' prefix followed by a descriptive verb-noun pair (health_check, web_search, x_search). The naming is uniform and predictable, aiding agent selection.
With only 3 tools, the server is lean but covers its core functionality adequately. The count is on the lower end of the ideal range but not egregiously small, as each tool serves a distinct and necessary role.
The tool set covers health verification and two main search domains (web and X), which seems sufficient for its stated purpose. Minor gaps like additional search refinement or account management are absent but not critical for basic operation.
Maintenance
Related MCP Connectors
Hosted X (Twitter) MCP server for Claude, ChatGPT and Cursor, with each action switched per account.
Official MCP server for twitterapis.com. Twitter/X search, users, DMs, Grok, monitoring and writes.
Hosted MCP server from TwitterAPI.io that gives AI agents read-only access to Twitter/X data. 12 tools: tweet search with X operators, user profiles, a user's latest tweets, followers and followings, replies, quotes, retweeters, mentions and trends. Connect at https://mcp.twitterapi.io/mcp with a TwitterAPI.io API key or account sign-in; usage is paid per result returned. Separately, TwitterAPI.io also offers x402 pay-per-call HTTP endpoints (tweet search, user info, a user's latest tweets) paid in USDC on Base or Solana with no account or API key; see https://docs.twitterapi.io/x402-pay-per-call. The MCP server itself uses an API key or sign-in.
Related MCP Servers
- AlicenseBqualityDmaintenanceMCP server for interacting with the X platform (Twitter) via MCP clients like Claude, Cursor AI, and Windsurf AI.206MIT
- AlicenseNot gradedqualityDmaintenanceMCP server that enables coding agents to search X/Twitter and the web through the local Grok CLI, providing better search results than default Codex or Claude search without requiring an API key.7MIT
- AlicenseNot gradedqualityCmaintenanceMCP server for xAI Grok API with 19 tools for chat, web and X search, session management, tool chaining, image and video generation, and documentation.2MIT
- AlicenseAqualityBmaintenanceMCP server for the Twitter/X API, enabling search, user profiles, tweets, followers, and more via natural language.94488 npm7MIT