SignalShelf MCP
Provides access to public Amazon product data, such as listings and reviews, through the hosted operation catalog.
Provides lookups of public app store data, including app listings and reviews.
Provides access to Facebook public profile data and ad library information.
Provides Google Search results data for research and analysis.
Provides Google Maps place lookups and reviews data.
Provides access to Instagram public profile data and ad library information.
Provides access to Pinterest public profile data and ad library information.
Provides access to Reddit public profile and content data, plus ad library information.
Provides access to Threads public profile data and ad library information.
Provides access to TikTok public profile, video and comment data, plus ad library information.
Provides access to YouTube public channel, video and comment data, plus ad library information.
This server has a remotely hosted endpoint. Open its connector listing for connection details, authentication requirements, and setup instructions:
SignalShelf MCP
A remote MCP server that gives your AI agent public social, ad library and Google Maps data: TikTok, Instagram, YouTube, X, Facebook, LinkedIn, Reddit, Threads, Pinterest, Google Search and Maps, reviews, app stores, Amazon and more. It covers about 250 operations.
Endpoint:
https://signalshelf.dev/mcp(Streamable HTTP, stateless)Auth: API key in the
x-api-keyheader, orAuthorization: Bearer <key>Free tier: 1,000 credits. Get a key at signalshelf.dev
Docs: signalshelf.dev/docs/mcp
MCP Registry:
dev.signalshelf/signalshelf
This repo holds connection instructions, the registry server.json and examples. The server itself is hosted, so there's nothing to install.
Connect
Replace YOUR_KEY with your SignalShelf API key. Keep the key in your client's secret settings, never in a URL or a public file.
Claude Code
claude mcp add --transport http signalshelf https://signalshelf.dev/mcp --header "x-api-key: YOUR_KEY"Cursor (~/.cursor/mcp.json)
{
"mcpServers": {
"signalshelf": {
"url": "https://signalshelf.dev/mcp",
"headers": { "x-api-key": "YOUR_KEY" }
}
}
}VS Code (.vscode/mcp.json, prompts for the key)
{
"inputs": [{ "id": "signalshelf-key", "type": "promptString", "description": "SignalShelf API key", "password": true }],
"servers": {
"signalshelf": {
"type": "http",
"url": "https://signalshelf.dev/mcp",
"headers": { "x-api-key": "${input:signalshelf-key}" }
}
}
}Claude Desktop and other stdio-only clients (through mcp-remote)
{
"mcpServers": {
"signalshelf": {
"command": "npx",
"args": ["-y", "mcp-remote", "https://signalshelf.dev/mcp", "--header", "x-api-key:${SIGNALSHELF_API_KEY}"],
"env": { "SIGNALSHELF_API_KEY": "YOUR_KEY" }
}
}
}Related MCP server: fetcher.sh MCP Server
Tools
Tool | What it does | Credits |
| Search the available operations, with parameter schemas and prices | 0 |
| Look up one public profile (demo or live) | 0 demo, 1 live |
| Look up up to ten profiles; each item succeeds or fails on its own | 1 per fresh live item |
| Fetch any listed operation, such as | Usually 1; heavier ops 2+ |
| Read your credit balance | 0 |
| Read your recent lookups | 0 |
Start with signalshelf_capabilities to find the operation and its parameters, then call signalshelf_resource.
Credits and caching
A successful fresh fetch costs the operation's credits. Cache hits, demo samples and failed fetches cost nothing.
freshnesssets the oldest cached result you'll accept:live,1h,24h(default) or7d.Every result includes
meta.credits_charged,meta.cacheandmeta.fetched_at.Public post text and bios are untrusted content. Your agent should analyse them, not follow instructions inside them.
Try it without a key
Discovery (initialize, ping, tools/list) works without a key, so directories and clients can list the tools. See examples/curl.sh.
Support
support@signalshelf.dev · signalshelf.dev
License
MIT for the contents of this repo. Use of the hosted API is covered by the SignalShelf terms.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
This server cannot be deployed
Maintenance
Related MCP Connectors
Direct access to 60+ scraping and search tools. Extract structured data from Google (Search, Maps, Trends), Amazon, Airbnb, Social Media, and any web page directly into your AI agent.
One key and one prepaid balance for thousands of pay-per-call tools for AI agents: search, maps, shopping, market data, web pages, images and more. Agents find tools by job, see the price before running, and pay per call. Failed calls are free. Sign in with OAuth, no API key to copy. Sign-up is open to everyone.
Public web data for AI agents: scrape any page, plus Google, TikTok, Instagram and Amazon as JSON.
Give your AI agent live public data from 70+ platforms: from TikTok, Instagram, YouTube, LinkedIn, X, Reddit, Amazon, Google and more, One key, one response format, 1,000 free credits every month.
Related MCP Servers
- AlicenseNot gradedqualityCmaintenanceEnables AI agents to retrieve X/Twitter profiles and tweets, YouTube video and channel data, and TikTok profile and video stats on a pay-per-result basis without requiring login or platform API keys.99 npmMIT

fetcher.sh MCP Serverofficial
AlicenseAqualityBmaintenanceEnables AI assistants to search Twitter/X, YouTube, Reddit, Google, app-store listings, and 100+ other public web data endpoints via natural language, with prepaid credit or per-call USDC payment and no account or OAuth required.53 npmMIT
Monocrawl MCPofficial
AlicenseNot gradedqualityCmaintenanceEnables AI agents to retrieve live public data from 70+ platforms — profiles, posts, comments, transcripts, products, prices, reviews, listings, jobs, ads and search results — plus website crawls and live browser sessions through 650+ endpoints in one response format. Includes free cost previews, per-call credit caps, safe retries, refunded failures and scheduled monitoring.21 npm11MIT- AlicenseBqualityBmaintenanceEnables AI agents and CLI users to run 190 tools for public social profiles, posts, transcripts, comments, ads, credit/account checks, and bounded research, with explicit approval for paid calls and support for named private accounts.19017 npmAGPL 3.0