Live X/Twitter search + AI trend digests for MCP clients. No Twitter API key — pay per call in USDC via x402.
- ✓Open-source license (MIT)
- ✓Actively maintained (<30d)
- ✓Clear description
- ✓Documented (README)
git clone https://github.com/ciinkwia/twitter-signal-mcp{
"mcpServers": {
"twitter-signal-mcp": {
"command": "node",
"args": ["/path/to/twitter-signal-mcp/dist/index.js"]
}
}
}Resumen de MCP Servers
# twitter-signal-mcp
Live **X/Twitter search** and **AI trend digests** inside Claude, Cursor, or any MCP client —
**no Twitter/X API key, no subscription, no monthly minimum.** Your agent pays a fraction of a
cent per call in USDC on Base using [x402](https://x402.org).
```
x_search $0.02/call up to 20 newest matching tweets + engagement
x_digest $0.05/call the same tweets + an AI-written trend summary
x_leads $0.05/call the people behind matching posts, enriched (bio, website, emails, followers)
x_pain_points $0.05/call clustered complaints / feature requests / praise for a product, with quotes
x_user_timeline $0.02/call one account's own recent posts
x_profile $0.02/call one account's bio, stats, and links
x_tweet $0.01/call one tweet by ID, full text + engagement
x_replies $0.02/call the reply thread under one tweet
x_find_accounts $0.02/call accounts matching a name/topic, not posts
x_account_verdict $0.02/call AI read on whether an account is worth engaging
x_watch $0.05/call only the posts newer than your last check
```
## Why this exists
The official X API starts at a monthly subscription and a developer-account approval. If an agent
just needs to know *what people are saying right now*, that's a lot of setup for a handful of
queries. This server charges per request instead — a hundred searches costs about sixty cents,
and there is nothing to cancel.
## Tools
### `x_search` — $0.02
Full advanced-search syntax. Returns up to 20 newest matches with author handle, follower count,
text, timestamp, permalink, likes, retweets, replies, and views. **Need more than 20?** A full
answer ends with `next.older` — pass its `cursor` value back (same `query`) for the next 20 older
matches, $0.02 a page. No slicing dates by hand.
### `x_digest` — $0.05
Everything above **plus** an AI-written digest generated only from the returned tweets:
```json
{
"summary": "3-5 sentences on what's actually being said",
"sentiment": "positive | negative | mixed | neutral",
"sentiment_note": "why it leans that way",
"key_themes": ["..."],
"driving_accounts": ["@handle", "..."],
"notable_post_urls": ["https://x.com/..."]
}
```
Use `x_digest` when you want the takeaway; `x_search` when you want the raw posts.
### `x_leads` — $0.05
Same advanced-search syntax as `x_search`, but returns the PEOPLE behind the matching posts
instead of the posts themselves — enriched with bio, website, emails found in the bio, follower
count, location, a DM-open flag, and every matching post per person:
```json
{
"result_count": 20,
"lead_count": 8,
"leads": [{ "username": "someone", "followers": 5000, "emails": [], "matched_posts": [] }]
}
```
Inputs: `query` (required, same syntax as `x_search`), `max` (optional, `"20"` or `"40"` — default 20).
Use for recruiting, cofounder search, B2B lead generation, influencer discovery, community sourcing.
### `x_pain_points` — $0.05
Pass a product or brand name/handle, get clustered complaints, feature requests, and praise mined
from live X posts, with verbatim quotes and URLs:
```json
{
"product": "Cursor",
"analysis": { "summary": "...", "pain_points": [{ "theme": "...", "count": 4, "severity": "medium" }] }
}
```
Input: `product` (required — a name or an X handle). Use for product research, competitor research,
churn signals, roadmap input.
### `x_user_timeline` — $0.02
An account's own recent posts (originals, replies, retweets), newest first. Input: `username`
(required). Use: `x_user_timeline username="elonmusk"` to track a competitor's or founder's own feed.
### `x_profile` — $0.02
One account's bio, website, location, join date, follower/following counts, and verified status.
Input: `username` (required). Use: `x_profile username="cursor_ai"` to qualify a lead before you
reach out.
### `x_tweet` — $0.01
One tweet by numeric ID: full text, author, timestamp, permalink, engagement. Input: `id` (required).
Use: `x_tweet id="1859012345678901234"` to re-check a specific post you already found.
### `x_replies` — $0.02
The reply thread under one tweet, newest first. Input: `id` (required, same as `x_tweet`). Use:
`x_replies id="1859012345678901234"` to see how people reacted to a specific post.
### `x_find_accounts` — $0.02
Search for ACCOUNTS matching a name, keyword, or topic — not posts. Input: `query` (required, same
syntax as `x_search`). Use: `x_find_accounts query="solar installer hawaii"` to build a shortlist of
accounts to follow up on.
### `x_account_verdict` — $0.02
A short AI-written read on one account: what it's about, how active/engaged it is, and whether it's
worth following up with. Input: `username` (required). Use: `x_account_verdict username="someone"`
to triage a lead fast without reading the profile yourself.
### `x_watch` — $0.05
Same search as `x_search`, but only returns posts newer than a cursor you supply. Inputs: `query`
(required), `since` (optional — an ISO timestamp or the `cursor.newest_id` from your last `x_watch`
call). Use: poll a brand/ticker/hashtag on a schedule without re-paying for posts you've already seen.
**A check that finds nothing new is not charged** — you only pay when there is news.
### Query syntax (`x_search`, `x_digest`, `x_leads`, `x_find_accounts`, `x_watch`)
Both tools take one `query` string:
| Pattern | Example |
|---|---|
| Keywords / phrases | `"rate cut" inflation` |
| Account | `from:elonmusk starship` |
| Hashtag / cashtag | `#bitcoin` · `$NVDA` |
| Engagement floor | `min_faves:100` · `min_retweets:50` |
| Language / dates | `lang:en since:2026-08-01 until:2026-08-09` |
| Exclusions | `airdrop -giveaway` |
## Setup
You need a Coinbase CDP wallet funded with a little USDC on Base — that wallet pays per call.
Create one at [portal.cdp.coinbase.com](https://portal.cdp.coinbase.com) (API key + wallet secret).
**Claude Desktop** — add to `claude_desktop_config.json`:
```json
{
"mcpServers": {
"twitter-signal": {
"command": "npx",
"args": ["-y", "twitter-signal-mcp"],
"env": {
"CDP_API_KEY_ID": "your-key-id",
"CDP_API_KEY_SECRET": "your-key-secret",
"CDP_WALLET_SECRET": "your-wallet-secret",
"X402_MAX_PRICE": "0.05"
}
}
}
}
```
`X402_MAX_PRICE` is a hard per-call ceiling in USD. If the endpoint ever quotes more than this,
the request is refused **before** anything is signed — no charge. The default `0.05` already covers
every tool above (the priciest are `x_digest`, `x_leads`, `x_pain_points`, and `x_watch` at $0.05) —
you don't need to raise it.
## Cost & safety
- You are only ever charged for a successful response. Invalid queries are rejected **before**
payment (HTTP 400, no settlement).
- **Nothing found = nothing charged.** An empty `x_search` / `x_digest` / `x_leads` /
`x_pain_points` / `x_watch` comes back as `no_results` with `charged: false` — the payment is
never settled.
- **Same call within 10 minutes = free.** Repeat the exact call from the same wallet and you get
the saved answer (`meta.charged: false`) at no cost.
- If the upstream tweet source fails after payment, the API returns an explicit `error` and logs
the call for refund rather than pretending it worked.
- Gas is covered by the facilitator — you spend USDC only.
- Identical queries within 60 seconds are served from cache.
## How it works
```
MCP client → twitter-signal-mcp → 402 challenge → sign USDC authorization
→ facilitator settles on Base → data returned
```
Protocol: **x402 v2** (CAIP-2 networks, `PAYMENT-REQUIRED` challenge header). A v1 client cannot
talk to this API.
## License
MIT
Lo que la gente pregunta sobre twitter-signal-mcp
¿Qué es ciinkwia/twitter-signal-mcp?
+
ciinkwia/twitter-signal-mcp es mcp servers para el ecosistema de Claude AI. Live X/Twitter search + AI trend digests for MCP clients. No Twitter API key — pay per call in USDC via x402. Tiene 1 estrellas en GitHub y su última actualización registrada es del 2026-09-20.
¿Cómo se instala twitter-signal-mcp?
+
Puedes instalar twitter-signal-mcp clonando el repositorio (https://github.com/ciinkwia/twitter-signal-mcp) o siguiendo las instrucciones del README en GitHub. ClaudeWave también te ofrece bloques de instalación rápida en esta misma página.
¿Es seguro usar ciinkwia/twitter-signal-mcp?
+
Nuestro agente de seguridad ha analizado ciinkwia/twitter-signal-mcp y le ha asignado un Trust Score de 87/100 (tier: Trusted). Revisa el desglose completo de comprobaciones superadas y flags en esta página.
¿Quién mantiene ciinkwia/twitter-signal-mcp?
+
ciinkwia/twitter-signal-mcp es mantenido por ciinkwia. La última actividad registrada en GitHub es del 2026-09-20, con 0 issues abiertos.
¿Hay alternativas a twitter-signal-mcp?
+
Sí. En ClaudeWave puedes explorar mcp servers similares en /categories/mcp, ordenados por popularidad o actividad reciente.
Despliega twitter-signal-mcp en tu cloud
Lleva este repo a producción en minutos. Cada plataforma genera su propio entorno con variables de entorno editables.
¿Mantienes este repo? Añade un badge a tu README
Pega el badge en tu README de GitHub para mostrar que está auditado por ClaudeWave. Cada badge enlaza de vuelta a esta página y muestra el Trust Score actual.
[](https://claudewave.com/repo/ciinkwia-twitter-signal-mcp)<a href="https://claudewave.com/repo/ciinkwia-twitter-signal-mcp"><img src="https://claudewave.com/api/badge/ciinkwia-twitter-signal-mcp" alt="Featured on ClaudeWave: ciinkwia/twitter-signal-mcp" width="320" height="64" /></a>Más MCP Servers
Fair-code workflow automation platform with native AI capabilities. Combine visual building with custom code, self-host or cloud, 400+ integrations.
User-friendly AI Interface (Supports Ollama, OpenAI API, ...)
An open-source AI agent that brings the power of Gemini directly into your terminal.
Real-time global intelligence dashboard. AI-powered news aggregation, geopolitical monitoring, and infrastructure tracking in a unified situational awareness interface
🕷️ An adaptive Web Scraping framework that handles everything from a single request to a full-scale crawl! Don't be shy, join here: https://discord.gg/EMgGbDceNQ
The fastest path to AI-powered full stack observability, even for lean teams.