{
  "markdown": "# BPE MCP Server\n\nModel Context Protocol server for the [BPE](https://bxetech.com) market-data\nAPI. Lets Claude Desktop, Cursor, Goose, and other MCP-enabled agents query\nconsolidated Bitcoin pricing, ML signals, funding-rate skew, and\nnatural-language market briefings — without writing any HTTP boilerplate.\n\n> **Status:** v0.2 — five tools, local stdio transport, no LLM\n> calls inside the briefing endpoint (template-rendered). API key required.\n\nListed in the [official MCP Server Registry](https://registry.modelcontextprotocol.io/v0.1/servers?search=io.github.bxetech/bpe-mcp) as `io.github.bxetech/bpe-mcp`. Install via `npm install -g @bxetech/bpe-mcp` or `npx -y @bxetech/bpe-mcp`.\n\n## Why use it\n\nIf you're building an agent that touches Bitcoin markets, you have two\noptions today:\n\n1. **Talk to 36 different exchange APIs directly.** Each has a different\n   schema, different auth, different rate limits, different reliability.\n   Your agent burns tokens reasoning over inconsistent payloads.\n2. **Use one consolidated source.** BPE aggregates all 36 venues into a\n   single normalised feed with derived signals (ML predictions,\n   sentiment, funding skew, basis). One auth, one schema, one tool call.\n\nThis MCP server is option 2 dressed up as native tools your agent can\ndiscover and call.\n\n## Tools (v0.2)\n\n| Tool | What it does |\n|---|---|\n| `get_consolidated_price` | Current BTC price aggregated across the BPE network |\n| `get_market_briefing` | ~4 lines (brief) or 6–8 (detailed) natural-language summary — price, funding skew, ML signal, sentiment, anomalies |\n| `get_funding_skew` | Per-venue annualised funding rates + max-spread pair |\n| `get_ml_signal` | Current ML prediction (direction, probability, confidence) at the requested horizon |\n| `get_sentiment_snapshot` | Crypto Fear & Greed index, news sentiment, mempool stress |\n\n## Install\n\n```bash\nnpm install -g @bxetech/bpe-mcp\n```\n\nOr use `npx` with no install (recommended for Claude Desktop):\n\n```jsonc\n// ~/Library/Application Support/Claude/claude_desktop_config.json\n{\n  \"mcpServers\": {\n    \"bpe\": {\n      \"command\": \"npx\",\n      \"args\": [\"-y\", \"@bxetech/bpe-mcp\"],\n      \"env\": {\n        \"BPE_API_KEY\": \"bpe_...\"\n      }\n    }\n  }\n}\n```\n\nRestart Claude Desktop. The five tools appear under the MCP server picker.\n\n## Configuration\n\n| Env var | Default | Notes |\n|---|---|---|\n| `BPE_API_KEY` | *(required)* | Get one at https://bxetech.com/contact |\n| `BPE_BASE_URL` | `https://mcp.bxetech.com` | Override for self-hosted / local dev |\n| `BPE_TIMEOUT_MS` | `5000` | Per-request timeout |\n\n## Local development\n\n```bash\ngit clone https://github.com/bxetech/bpe-mcp\ncd bpe-mcp\nnpm install\nnpm run build\nBPE_API_KEY=... npm run dev\n```\n\nTo dogfood in Claude Desktop while iterating, point at the local build:\n\n```jsonc\n{\n  \"mcpServers\": {\n    \"bpe-dev\": {\n      \"command\": \"node\",\n      \"args\": [\"/abs/path/to/mcp-server/dist/index.js\"],\n      \"env\": {\n        \"BPE_API_KEY\": \"...\",\n        \"BPE_BASE_URL\": \"http://localhost:8082\"\n      }\n    }\n  }\n}\n```\n\n## Roadmap\n\n- **v0.2** — `get_sentiment_snapshot` ✅, friendly tier-aware errors ✅, timestamped briefings ✅\n- **v0.3** — `get_basis`, `subscribe_alert` (webhook subscription), hosted HTTP transport for non-developer install (Cursor extension, ChatGPT Desktop)\n- **v0.4** — Optional LLM-rendered briefing via Haiku (richer prose, with per-call cost)\n\nSee [`docs/research/AGENT_INTEGRATION_AND_MONETISATION_2026-04-27.md`](../docs/research/AGENT_INTEGRATION_AND_MONETISATION_2026-04-27.md) for the broader product / monetisation plan.\n\n## License\n\nMIT — see [`LICENSE`](./LICENSE).\n",
  "bytes": 3620,
  "sha": "47cafc19f2c526368a8b4c55c99dae022e062912cd936ff4a428adf11f31f642",
  "repo_slug": "bxetech/bpe-mcp",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/mcp_io_github_bxetech_bpe_mcp_e8d8d848/readme"
}