About SignalStack
SignalStack turns on-chain and market data into a daily crypto market Candle Review, available through a REST API and MCP for AI agents. Its 0–100 score covers seven categories, with category details available to Pro users.
Key features
- 01Publish a daily market score
- 02Combine seven categories into a score
- 03Retrieve historical reviews
- 04Query individual category breakdowns
- 05Offer REST and MCP access
Use cases
- 01Check the current crypto market stance
- 02Review historical market scores
- 03Examine signals in a specific category
MCP support
SignalStack provides MCP tools for today's score, historical reviews, and category breakdowns over SSE. Category details require Pro access, with an X-API-Key supplied on connection.
Configuration
Remote servers need no local install. Pick your client, then copy the config or install in one click.
https://mcp.sigintstack.io/mcpclaude mcp add --transport sse ud4090v-signalstack https://mcp.sigintstack.io/mcp{"mcpServers":{"ud4090v-signalstack":{"type":"sse","url":"https://mcp.sigintstack.io/mcp"}}}Open in Cursor{"servers":{"ud4090v-signalstack":{"type":"sse","url":"https://mcp.sigintstack.io/mcp"}}}Open in VS Code{"mcpServers":{"ud4090v-signalstack":{"type":"sse","url":"https://mcp.sigintstack.io/mcp"}}}{"mcpServers":{"ud4090v-signalstack":{"serverUrl":"https://mcp.sigintstack.io/mcp"}}}Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.
https://mcp.sigintstack.io/mcpAuth requiredTools
Details
From the project READMESignalStack
Daily crypto market intelligence — a 0-100 Candle Review score from real on-chain and market data. Available as a REST API and MCP server for AI agents.
What It Does
Every day at ~06:00 UTC, SignalStack pulls live data from DefiLlama, CoinGecko, and Dune Analytics, scores market conditions across 7 categories, and publishes a structured Candle Review.
Categories and weights:
| Category | Max Score |
|---|---|
| Price Action | 25 |
| On-Chain | 20 |
| Derivatives | 20 |
| ETF Flows | 15 |
| Macro | 10 |
| Dominance | 5 |
| Sentiment | 5 |
| Total | 100 |
Score labels:
- 80-100: Strongly Bullish
- 60-79: Bullish
- 40-59: Neutral
- 20-39: Bearish
- 0-19: Strongly Bearish
REST API — Live
Base URL: http://96.126.106.225:3458
OpenAPI spec: GET /openapi.yaml
Endpoints
| Endpoint | Tier | Description |
|---|---|---|
GET /health | Free | Health check |
GET /candle/today | Free + Pro | Today's Candle Review |
GET /candle/history?limit=30 | Free + Pro | Historical reviews |
GET /candle/category/{category} | Pro only | Category breakdown |
GET /openapi.yaml | Free | OpenAPI 3.0 specification |
Freemium Tiers
| Free | Pro | |
|---|---|---|
| API Key | Not required | X-API-Key header |
/candle/today | date, score, label | Full JSON |
/candle/history | date, score, label | Full JSON |
/candle/category/:cat | ❌ 403 | ✅ Full breakdown |
Getting an API Key
Contact the SignalStack team or check the RapidAPI listing for pro tier access. Free tier requires no key.
Example — Free tier
curl http://96.126.106.225:3458/candle/today
{
"date": "2026-03-28",
"score": 67,
"label": "Bullish"
}
Example — Pro tier
curl -H "X-API-Key: your-key-here" http://96.126.106.225:3458/candle/today
{
"date": "2026-03-28",
"score": 67,
"label": "Bullish",
"categories": {
"price_action": { "score": 18, "max": 25, "summary": "BTC up 3.2% on strong volume." },
"on_chain": { "score": 14, "max": 20, "summary": "Exchange reserves declining." },
"derivatives": { "score": 13, "max": 20, "summary": "Funding rates slightly positive." },
"etf_flows": { "score": 11, "max": 15, "summary": "Spot ETF net inflows $400M." },
"macro": { "score": 6, "max": 10, "summary": "DXY weakening, risk-on." },
"dominance": { "score": 3, "max": 5, "summary": "BTC dominance stable at 54%." },
"sentiment": { "score": 2, "max": 5, "summary": "Fear & Greed index at 68." }
},
"outlook": "Near-term momentum remains bullish.",
"what_happened": "Spot ETF inflows hit $400M. BTC held above $70k.",
"generated_at": "2026-03-28T06:00:01.000Z"
}
MCP Server — AI Agent Integration
SignalStack exposes an MCP (Model Context Protocol) server via SSE transport for direct AI agent integration.
Connection
SSE endpoint: GET http://96.126.106.225:3458/mcp
Request endpoint: POST http://96.126.106.225:3458/mcp
Available MCP Tools
| Tool | Tier | Description |
|---|---|---|
candle.get_today | Free + Pro | Today's composite score |
candle.get_history | Free + Pro | Historical reviews |
candle.get_category | Pro | Category breakdown |
Connecting with an MCP client
// Example using MCP SDK
const client = new Client({ name: "my-agent", version: "1.0.0" }, { capabilities: {} });
const transport = new SSEClientTransport(new URL("http://96.126.106.225:3458/mcp"));
await client.connect(transport);
const result = await client.callTool({ name: "candle.get_today", arguments: {} });
Pass X-API-Key as a header on the SSE connection for pro tier access.
OpenAPI Specification
The full OpenAPI 3.0 spec is served live at:
GET http://96.126.106.225:3458/openapi.yaml
Also available as a static file at openapi.yaml in the project root.
Stack
- Runtime: Node.js + TypeScript
- Framework: Express
- Data sources: DefiLlama (free), CoinGecko (free/demo), Dune Analytics
- Storage: SQLite (score store, api-keys.db)
- Transport: REST + MCP SSE
- Deploy: systemd service + nginx reverse proxy
Directory Listings
SignalStack is listed on:
- RapidAPI — REST API discovery
- Smithery — MCP server registry
- mcp.run — MCP server registry
- Futurepedia — AI tools directory
- There's An AI For That — AI tools directory
Versions
| Phase | Status | Description |
|---|---|---|
| P0-P2 | ✅ Complete | MCP tools: DefiLlama + CoinGecko + Dune (21 tools) |
| P3 | ✅ Complete | Candle Review scoring engine |
| P4 | ✅ Complete | REST API + public MCP SSE + freemium gating |
| P5 | ✅ Complete | OpenAPI spec + directory listings |
| P6 | Backlog | Stripe billing integration |
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.