Skip to main content
Glama
brssignals

BRS Signals MCP Server

Official

₿RS Signals — MCP Server

mcp-name: io.github.brssignals/brs-signals-mcp

Live Bitcoin market regime for AI agents: three independent sensors read pre-price data — fee curves, funding velocity, whale flows, absence — every 30s and reject almost everything. When they converge, you get a directional call with the evidence attached. The data can't be reconstructed from history. Provable silence: the public rejection funnel shows exactly how many reads died at each gate, so your agent can audit the silence, not just the signals. MCP + REST + x402. Free tier, no card.

Don't trust us — query us

The full public track record is an endpoint, not a PDF. Point your agent at it and let it reach its own conclusion:

"BRS Signals claims it rejects almost every read. Pull its public track
record (brs_audit_track_record) and today's rejection funnel
(brs_rejection_funnel), and tell me: does the data support the claim,
and what happened after each call at +4h and +24h?"

Both of those tools are keyless — no API key, no payment, no signup. That is the point: the honesty surface is open by design.

Related MCP server: fathom-fyi

Use it in 30 seconds — remote, no install

The public endpoint is live at https://brs-signals.com/mcp (streamable-http · keyless free tier). Any MCP client can use it — no package install, no local process.

Cursor (MCP requires Cursor's paid Pro tier) — Settings → Features → MCP → + Add New MCP Server → Type http, URL https://brs-signals.com/mcp. Or add .cursor/mcp.json:

{
  "mcpServers": {
    "brs-signals": { "type": "http", "url": "https://brs-signals.com/mcp" }
  }
}

Claude Desktop (MCP is free) — Settings → Developer → Edit Config:

{
  "mcpServers": {
    "brs-signals": { "type": "http", "url": "https://brs-signals.com/mcp" }
  }
}

Then just ask your agent: "What's the current Bitcoin regime and convergence score?"

Or install locally (stdio, self-host)

pip install brs-signals-mcp
export BRS_API_KEY=va_yourkey_here   # get one at https://brs-signals.com
brs-mcp                                # starts the MCP server (stdio)

Claude Desktop (stdio) — add to ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows), then restart:

{
  "mcpServers": {
    "brs-signals": {
      "command": "brs-mcp",
      "env": { "BRS_API_KEY": "va_yourkey_here" }
    }
  }
}

Cursor (stdio) — Settings → Features → MCP → Add New MCP Server: Type command · Command python3 -m mcp_brs · Env BRS_API_KEY.

What You Get

Three independent sensors running every 30 seconds, fused into one convergence score:

Sensor

Domain

What It Detects

X-Ray

On-chain

Bitcoin's internal structure — fee curves, capital movement, network health

Pulse

Off-chain

Derivatives positioning and macro — funding velocity, whale flows, squeeze probability

Shadow

Absence

What has stopped happening — the silence when a normally busy channel goes quiet

When all three agree → conviction. When they disagree → silence.

MCP Tools

7 tools. Discovery is authorization-scoped: keyless callers see only the 4 Free tools; a caller presenting a key sees all 7. No Pro tool serves data keyless — the metered tool below surfaces its x402 payment challenge keyless, but never data.

New to this server? Call brs_market_state (free) first — it shows the current regime, three-eye convergence, and system health. Need the directional posture one time? Call brs_decision_context with no key, read error.payment from the PAYMENT_REQUIRED result, pay, and re-call with the tx_signature.

Free — no key required

Tool

What it returns

brs_market_state

Regime structure + three-eye convergence + system health (call first)

brs_audit_track_record

Public proof — every call and what BTC did next (+4h/+24h)

brs_rejection_funnel

Per-gate cycle counts: why no signal came out

brs_system_status

Component health + sample size behind every reading

Pro — key or x402 required

Tool

What it returns

brs_decision_context

Directional posture (bullish / bearish / WAIT) with evidence

brs_history

Recent calls with resolved outcomes

brs_raw_stream

One raw stream: fees, funding, stablecoin, or gamma

WAIT means no edge. The system tells you which game the market is playing and how much the sensors agree — it is context, not an instruction to trade.

Pricing

  • Free tier — Regime classification + convergence score + the public proof (5 req/min). No card.

  • Pro tier ($50/mo) — Directional posture + raw streams + full history, 60 req/min.

  • Per-call ($0.01/query) — the Pro posture one query at a time via x402 (USDC on Solana or Base): hit /api/v2/bias/per-call, pay the 402, retry with ?tx_signature= + your free key. Capped at $50 per rolling 30 days, then converts to Pro — never more than $50 in any 30 days.

  • x402 — Agents with wallets pay per access in USDC; no signup, no human.

Get your API key at brs-signals.com.

Related MCP Connectors

Related MCP Servers

  • A
    license
    Not graded
    quality
    C
    maintenance
    AI-powered crypto signal intelligence for 20 assets (BTC, ETH, SOL, etc). 6 scoring dimensions: whale activity, technical analysis, derivatives flow, narrative strength, sentiment, market structure. Market regime detection (TRENDING/RANGING), portfolio optimization, and accuracy tracking. 9 read-only MCP tools. Free via MCP, $0.001 USDC via x402 on Base for REST API.
    3
    MIT
  • A
    license
    A
    quality
    D
    maintenance
    Financial intelligence for AI agents. 31 tools across 8 data sources — regime, derivatives, stablecoin flows, momentum, volatility, macro, DeFi, weather patterns, political cycles, seasonality. The context layer between your agent and a bad trade.
    31
    9 npm
    9
    MIT
  • A
    license
    Not graded
    quality
    B
    maintenance
    Classifies BTC market into 5 regimes (Risk-On, Range-Bound, Risk-Off, Capitulation, Euphoria) with target exposure and posture, enabling risk management decisions.
    MIT
  • A
    license
    Not graded
    quality
    C
    maintenance
    Provides AI agents with real-time financial market intelligence including regime detection, adaptive signals, macro context chains, and cross-market analysis for crypto, US, and Korean stocks.
    MIT