Context7
Up-to-date docs for AI agents
作者提供的说明(英文原文)
11 tools: crypto signals, sentiment, macro regime, arena, search, exec. x402 pay-per-call.
这三项永远不会合并成一个分数:意见、测量和人气是不同的东西。
我们还没有为它写通俗的介绍。下面是我们从注册中心读取到的事实。
选择你使用的 AI 应用,然后复制设置。
SIGNALFUSE_CREDIT_TOKEN标为机密的值是你账户的钥匙。千万不要分享给别人,也不要粘贴到聊天里。
你需要
你用的是哪个应用?
在上方选择你的应用。我们会记住它,下次按钮会显示“安装到 …”。
打开 Claude Desktop → Settings → Developer → Edit Config,把下面的内容粘贴到 claude_desktop_config.json
{
"mcpServers": {
"signalfuse": {
"command": "npx",
"args": [
"-y",
"signalfuse-mcp"
],
"env": {
"SIGNALFUSE_CREDIT_TOKEN": "your-value"
}
}
}
}在终端中运行这条命令
claude mcp add signalfuse --env SIGNALFUSE_CREDIT_TOKEN=your-value -- npx -y signalfuse-mcp运行前请先填写空着的值。
粘贴到项目中的 .cursor/mcp.json,或 ~/.cursor/mcp.json
{
"mcpServers": {
"signalfuse": {
"command": "npx",
"args": [
"-y",
"signalfuse-mcp"
],
"env": {
"SIGNALFUSE_CREDIT_TOKEN": "your-value"
}
}
}
}浏览器会先询问是否打开应用,然后应用会显示设置并请你确认。密钥的值会留空,由你自己填写。
粘贴到项目中的 .vscode/mcp.json
{
"servers": {
"signalfuse": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"signalfuse-mcp"
],
"env": {
"SIGNALFUSE_CREDIT_TOKEN": "your-value"
}
}
}
}浏览器会先询问是否打开应用,然后应用会显示设置并请你确认。密钥的值会留空,由你自己填写。
这个服务器在你的电脑上运行,而 ChatGPT 只能连接互联网上的服务器。请选择其他应用。
你的选择通过 Cookie 保存在这台设备上,无需账户。
这些设置是根据作者在注册中心的条目生成的。如果 README 的写法不同,请以 README 为准。
它在你的电脑上运行;安装后,工具列表会显示在你的 AI 应用中。
我们检查的是任何人都能看到的信息;我们没有阅读代码。不适用的检查项不计入总数。 · 检查于 2026年11月6日
注册名称已通过 GitHub 账户 hypeprinter007-stack 验证,该账户拥有此仓库
我们还没有从 GitHub 读取这个仓库
我们还没有从 GitHub 读取这个仓库
它在你的电脑上运行,没有互联网端点
在你的电脑上运行,改用本地设置
它在你的电脑上运行;我们只能读取互联网端点的工具列表
与注册中心里这个分类中所有活跃的服务器比较。只有当发布者拥有该仓库时,星标数才计入。
它在 1 项适用的基础检查中通过了 1 项。我们没有阅读代码。在添加密钥或令牌之前,请先看看这个服务器要求哪些权限,并使用只具备所需权限的密钥。
你需要填写:SIGNALFUSE_CREDIT_TOKEN
它在你的电脑上运行。你的 AI 应用会在需要时启动它,数据不会经过作者的服务器(除非它自己调用在线服务)。
GitHub 仓库最后更新于–,最新版本(1.1.4)于2个月前进入注册中心。
支持 MCP 的应用,例如 Claude Desktop、Claude Code、Cursor 和 VS Code。你的电脑上还需要有 Node.js。
同一分类中有:Context7, Codebase Memory, Exa
同一分类中其他作者的服务器,按星标数排序。
如何看卡片:★ GitHub 星标i三大支柱:评价 · 检查 · 使用情况i更新 = 最新版本发布到注册中心的时间i
Up-to-date docs for AI agents
Codebase knowledge graph for AI agents — 162 languages, sub-ms queries, 99% fewer tokens.
Connect AI agents to Exa for web search, content fetching, and multi-step research.
按原文从作者的仓库复制。PopNx MCP 没有审查或背书其内容。HTML、图片和脚本已被移除。链接会离开本站。
MCP server and REST API for crypto trading signals, sentiment analysis, macro regime classification, strategy arena, web search, and sandboxed code execution. Designed for AI agents, LLMs, and autonomous trading systems.
https://api.signalfuse.cohttps://api.signalfuse.co/openapi.jsonhttps://api.signalfuse.co/.well-known/x402https://api.signalfuse.co/llms.txtnpx signalfuse-mcp
Or install globally:
npm install -g signalfuse-mcp
pip install signalfuse
| Method | Path | Description | Price |
|---|---|---|---|
| GET | /v1/signal/{symbol} | Fused directional signal with entry/TP/SL | $0.010 |
| GET | /v1/signal/batch | Signals for all assets in one call | $0.075 |
| GET | /v1/sentiment/{symbol} | ModernFinBERT sentiment analysis | $0.002 |
| GET | /v1/regime | Macro regime classification (risk-on/off/neutral) | $0.001 |
| GET | /v1/arena/leaderboard | Strategy Arena rankings | FREE |
| GET | /v1/arena/{strategy_id}/{symbol} | Signal from a specific arena strategy | $0.001 |
| GET | /v1/gateway/search/brave | Brave web search | $0.008 |
| POST | /v1/gateway/search/tavily | Tavily AI-powered web search | $0.012 |
| POST | /v1/gateway/search/fused | Brave + Tavily fused and deduplicated | $0.015 |
| POST | /v1/gateway/execute/e2b | Sandboxed code execution (Python/JS) | $0.005 |
| GET | /v1/pricing | Pricing information | FREE |
| GET | /v1/credits/balance | Credit balance check | FREE |
| POST | /v1/credits/trial | Claim 5 free trial credits | FREE |
Get 5 free credits with no signup. Send your Ethereum wallet address:
curl -X POST https://api.signalfuse.co/v1/credits/trial \
-H "Content-Type: application/json" \ -d '{"wallet": "0xYOUR_ETH_ADDRESS"}'
Response:
{
"credit_token": "sf_abc123...", "credits": 5, "wallet": "0xYOUR_ETH_ADDRESS" }
Pass the returned credit_token as the X-Credit-Token header on subsequent requests, or use it in MCP tool calls.
curl https://api.signalfuse.co/v1/signal/BTC \
-H "X-Credit-Token: sf_abc123..."
{
"symbol": "BTC", "signal": "long", "signal_strength": 74, "confidence": 0.82, "regime": "risk_on", "components": { "social": {"score": 0.49, "label": "bullish"}, "macro": {"score": 0.63, "label": "bullish"}, "market": {"score": 0.42, "label": "long_bias"} }, "updated_at": "2026-04-04T12:00:00Z" }
curl https://api.signalfuse.co/v1/sentiment/ETH \
-H "X-Credit-Token: sf_abc123..."
{
"symbol": "ETH", "score": 0.65, "label": "bullish", "model": "ModernFinBERT", "updated_at": "2026-04-04T12:00:00Z" }
curl https://api.signalfuse.co/v1/regime \
-H "X-Credit-Token: sf_abc123..."
{
"regime": "risk_on", "confidence": 0.78 }
curl https://api.signalfuse.co/v1/arena/leaderboard
[
{ "strategy_id": "ema_5_breakout", "wins": 142, "losses": 98, "win_rate": 0.591, "total_pnl_bp": 340, "total_income_usd": 1720.50 } ]
curl "https://api.signalfuse.co/v1/gateway/search/brave?q=bitcoin+ETF+flows" \
-H "X-Credit-Token: sf_abc123..."
curl -X POST https://api.signalfuse.co/v1/gateway/search/tavily \
-H "Content-Type: application/json" \ -H "X-Credit-Token: sf_abc123..." \ -d '{"query": "bitcoin ETF flows", "search_depth": "advanced"}'
curl -X POST https://api.signalfuse.co/v1/gateway/execute/e2b \
-H "Content-Type: application/json" \ -H "X-Credit-Token: sf_abc123..." \ -d '{"code": "print(2 + 2)", "language": "python"}'
{
"stdout": "4\n", "stderr": "", "exit_code": 0 }
| Tier | Credits | Price | Per-Credit Cost |
|---|---|---|---|
| Free Trial | 5 | $0 | $0 |
| x402 Pay-Per-Call | unlimited | market rate | see endpoint table |
| Starter Pack | 500 | available at signalfuse.co | bulk discount |
| Pro Pack | 5,000 | available at signalfuse.co | bulk discount |
x402 is an open protocol for HTTP micropayments. SignalFuse accepts USDC on Base. No API keys, no subscriptions -- each request includes a payment header and the server settles on-chain. See the x402 spec for details.
When connected as an MCP server, SignalFuse exposes 11 tools:
| Tool Name | Description |
|---|---|
get_signal | Fused directional signal for a crypto asset (direction, strength, confidence, regime, entry/TP/SL) |
get_signal_batch | Signals for all supported assets in one call |
get_sentiment | ModernFinBERT sentiment analysis (score, label) |
get_regime | Macro regime classification (risk_on, risk_off, neutral) |
get_arena_leaderboard | Strategy Arena rankings with win rate, P&L, income (free) |
get_arena_signal | Signal from a specific arena strategy for a given asset |
search_brave | Web search via Brave Search |
search_tavily | AI-powered web search via Tavily |
execute_code | Sandboxed Python/JavaScript execution via E2B |
get_pricing | Pricing information (free) |
check_balance | Credit balance check (free) |
| Strategy ID | Description |
|---|---|
ema_5_breakout | Trend-following on exponential moving average crossovers |
rsi_7_extremes | Momentum entries at RSI overbought/oversold levels |
rsi_reversion_swing | Mean-reversion fades against RSI extremes |
vwap_reversion | Mean-reversion entries around volume-weighted average price |
Add to ~/Library/Application Support/Claude/claude_desktop_config.json:
{
"mcpServers": { "signalfuse": { "command": "npx", "args": ["-y", "signalfuse-mcp"] } } }
Add to your project's .mcp.json:
{
"mcpServers": { "signalfuse": { "command": "npx", "args": ["-y", "signalfuse-mcp"] } } }
from langchain_signalfuse import SignalFuseTool, SentimentTool, MacroRegimeTool
tools = [ SignalFuseTool(credit_token="sf_abc123..."), SentimentTool(credit_token="sf_abc123..."), MacroRegimeTool(credit_token="sf_abc123..."), ] agent = initialize_agent(tools, llm, agent=AgentType.ZERO_SHOT_REACT_DESCRIPTION)
See langchain_signalfuse.py in this repository for the full tool set (7 tools).
import signalfuseActions from "./elizaos_plugin";
// Register actions: getSignalAction, getSentimentAction, getRegimeAction, getSignalBatchAction, getArenaLeaderboardAction
See elizaos_plugin.ts in this repository for the full plugin.
import requests
response = requests.get( "https://api.signalfuse.co/v1/signal/BTC", headers={"X-Credit-Token": "sf_abc123..."} ) signal = response.json() print(signal["signal"], signal["signal_strength"])
const res = await fetch("https://api.signalfuse.co/v1/signal/BTC", {
headers: { "X-Credit-Token": "sf_abc123..." } }); const signal = await res.json(); console.log(signal.signal, signal.signal_strength);
BTC, ETH, SOL, DOGE, PEPE, WIF, BONK, ARB, OP, AVAX -- with more added regularly. Use get_signal_batch with no arguments to retrieve the current full list.
io.github.hypeprinter007-stack/signalfusesmithery.yaml in this repositorySignalFuse is a data fusion API, not financial advice. Signals are mathematical composites derived from sentiment, macro, and market structure data. They can be wrong. Trade at your own risk.
MIT