# KK6BZB/signallord-mcp-server [Health: Active]

**Category:** 💰 Finance & Fintech  
**Repository:** https://github.com/KK6BZB/signallord-mcp-server  
**GitHub Stars:** 1  
**npm Downloads (last month):** 74  
**Views:** 4  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/kk6bzb-signallord-mcp-server

## Description
Bitcoin regime detection intelligence — know which market regime you are in and the probabilities of what happens next. 17 tools with historical pattern matching, Odin AI analysis, composite gauges, and real-time macro/institutional/on-chain data.

## Tools
Capabilities this server exposes over MCP:

- **signallord_get_regime** — Current regime, historical matches, outcome probabilities
- **signallord_get_regime_history** — Regime evolution over time from daily snapshots
- **signallord_get_composite_score** — Composite 0-100 score with regime label
- **signallord_get_gauge_breakdown** — Individual gauge scores (4 free, 8 premium)
- **signallord_get_gauge_history** — Historical scores for trend analysis
- **signallord_get_odin_analysis** — AI market analysis (basic free, enriched premium)
- **signallord_ask_odin** — Ask Odin a question about BTC markets
- **signallord_get_btc_metrics** — Price, volume, derivatives (+ on-chain on premium)
- **signallord_get_institutional_data** — ETF flows, Coinbase premium, CME basis
- **signallord_get_macro_indicators** — DXY, VIX, Fear & Greed, liquidity stack
- **signallord_get_what_changed** — Day-over-day market changes
- **signallord_get_btc_vix_correlation** — BTC vs VIX correlation + spike detection
- **signallord_get_btc_gold_ratio** — BTC vs Gold normalized comparison
- **signallord_get_etf_flow_history** — Daily ETF flow history (since Jan 2024)
- **signallord_get_exchange_flow_history** — Exchange flow accumulation/distribution
- **signallord_get_liquidation_data** — Liquidation momentum scoring
- **signallord_get_odin_track_record** — AI prediction accuracy and learned patterns
- **signallord_get_pi_cycle** — 111DMA vs 350DMA·2 cross — cycle-top warning
- **signallord_get_realized_price** — Spot vs on-chain cost basis + reclaim/loss events
- **signallord_get_reserve_risk** — Price/HODL Bank ratio banded into cycle phases
- **signallord_get_mtf_trend** — D/W/M trend alignment composite (-3 to +3)
- **signallord_get_hash_ribbons** — 30/60-day hashrate MA cross + 30d divergence z-score
- **signallord_get_capitulation_score** — Composite of volume, liquidations, funding, OI, velocity
- **signallord_get_cycle_bottom_probability** — Weighted accumulation playbook score (0-100)
- **signallord_get_cycle_top_probability** — Weighted distribution playbook score (0-100)
- **signallord_get_data_health** — Which signals are currently stale due to upstream data issues

## Claude Desktop Quick Installation
Install path detected from listing signals. Uses `npx` (confidence: high):

```json
"mcpServers": {
  "signallord-mcp-server": {
    "command": "npx",
    "args": ["-y","@modelcontextprotocol/inspector"],
    "env": {
      "SIGNALLORD_API_KEY": "",
      "SIGNALLORD_API_BASE_URL": "",
      "MCP_SERVER_PORT": ""
    }
  }
}
```

**Requires environment variables:** `SIGNALLORD_API_KEY`, `SIGNALLORD_API_BASE_URL`, `MCP_SERVER_PORT` — the values above are empty placeholders; fill in real credentials before running (see the repository for what each one is for).

## Documentation

## What KK6BZB/signallord-mcp-server MCP server does

KK6BZB/signallord-mcp-server MCP server gives AI agents access to Signal Lord’s Bitcoin regime and market-analysis features. Its tools identify the current market regime, compare present conditions with historical periods, and return probabilities for possible outcomes. The service also exposes composite scores, gauge histories, macro indicators, institutional data, derivatives metrics, exchange flows, liquidations, and several Bitcoin cycle indicators.

The toolset includes 26 operations grouped around regime detection, gauges, Odin AI, market data, signals, cycle indicators, probability playbooks, and data health. Examples include `signallord_get_regime`, `signallord_get_composite_score`, `signallord_get_macro_indicators`, `signallord_ask_odin`, and `signallord_get_data_health`. Some tools or fields return limited information on the free tier, while premium and pro tiers provide broader data access.

## How it works

The server acts as a stateless MCP protocol adapter. An AI client sends MCP tool requests to the local process, which makes HTTP requests to the configured Signal Lord backend. Authentication is passed through the `X-SIGNALLORD-API-KEY` header. Regime detection, historical matching, rate limiting, data collection, and Odin AI processing are handled by the backend rather than locally.

The default local transport is streamable HTTP, with the server listening on port 8100. Clients can connect through the local MCP endpoint at `/mcp`, or configure a local command in Claude Desktop. The README also identifies Claude and Cursor as example AI agents in the architecture.

## Setup and configuration

To run KK6BZB/signallord-mcp-server MCP server, install the published package globally with npm, and ensure Python 3.10 or newer and `pip` are available on the system. The package installs its dependencies automatically. The executable is `signallord-mcp`.

Create an account at Signal Lord and generate an API key. The key is shown only once and should be assigned to `SIGNALLORD_API_KEY`. `SIGNALLORD_API_BASE_URL` selects the backend URL; the documented production value is `https://signallord.onrender.com`, while the server default is `http://localhost:8000`. `MCP_SERVER_PORT` controls the local listening port and defaults to `8100`.

After starting the process, connect an MCP client to `http://localhost:8100/mcp`, or use a client configuration that launches `signallord-mcp` with the required environment variables.

## Limitations and notes

KK6BZB/signallord-mcp-server MCP server requires a Signal Lord API key and access to the configured backend. Availability and detail depend on the account tier: the documented free tier includes 10 requests per minute and 500 per day, while premium and pro tiers have higher limits and additional data.

The data-health tool can identify signals that are stale because of upstream data problems. This matters when an agent uses current gauges, market metrics, or derived indicators. The server provides analysis and historical probabilities, not a guarantee of future Bitcoin price movements. The project is licensed under MIT.

_Full upstream README: https://allmcps.com/mcp/kk6bzb-signallord-mcp-server/readme_