Trading Signals

io.github.roman-rr由 GitHub 帳號 roman-rr 發布v1.1.0更新於 Nov 2, 2026

Live AI crypto trading signals for agents. Try 3 signals without a key; sign in from your agent.

發布方自己的服務：SIGNALS Loyalty Agent Platform

已驗證Streamable HTTP可網頁執行金融 AI 與機器學習

服務詳細資訊

識別碼: io.github.roman-rr/trading-signals
傳輸: Streamable HTTP
執行環境: Web executable
信任: Verified
發布於: Sep 16, 2026

安全提示

MCP 服務會接收對話中的工具呼叫請求，並可能執行外部操作。啟用前請先了解該服務能做什麼，並只授予它所需的權限。

註冊中心已驗證此條目所屬的 GitHub 帳號 roman-rr。這並不表明該帳號就是產品方。

SourceWeft 的檢查只涵蓋條目是否完整、端點是否回應；不審查伺服器的程式碼，也不審查其工具的行為。

在 README 中新增徽章

向讀者展示該 MCP 服務已收錄於 SourceWeft，徽章連結到本頁。

[![Listed on SourceWeft](https://sourceweft.com/badge/mcp/io.github.roman-rr%2Ftrading-signals.svg)](https://sourceweft.com/mcp/io.github.roman-rr%2Ftrading-signals?utm_source=readme&utm_medium=badge)

概覽

AI 產生的概覽

遠端 MCP 伺服器，讓助理取得附帶推理依據、驗證結果與整體績效統計的即時 AI 加密貨幣交易訊號。

可以做什麼: 列出仍處於開放狀態的加密交易訊號，含幣種、方向、信心分數與背後推理
依 ID 取得單一訊號（進行中或已結算），含價格、損益以及是否觸及止盈或止損
檢視已結算訊號的歷史紀錄，依時間新到舊並附驗證結果
讀取命中率、已驗證數量、平均槓桿、方向分布等彙總統計，帶金鑰時另有 ROI 資料
呼叫已棄用的 register 工具，它只會回傳註冊連結
適用情境: 適合讓助理彙整目前的加密交易想法、說明訊號背後的推理，或回顧過去訊號的表現。偏向研究與模擬交易流程，而非自動下單執行。
執行需求: 遠端 HTTP 端點，不需本機執行環境或安裝套件。公開網址免帳號即可使用，回傳 3 筆訊號預覽。完整訊號、歷史紀錄與 ROI 統計需透過用戶端登入，或提供 API 金鑰：標頭 X-Api-Key、查詢參數 apiKey，或 Codex 使用的環境變數 SIGNALS_API_KEY。

安裝前請注意

訊號是第三方交易建議而非投資建議，README 中的範例列明確不是即時資料。Pro 功能（進場價、止損、止盈、槓桿、部位大小）在 14 天試用後為每月 35 美元或每年 300 美元，未訂閱則不扣款。把 API 金鑰放在網址查詢參數中可能經由日誌外洩。register 工具已棄用且不儲存任何內容。

安裝

在 SourceWeft 中

開啟儀表板中的 Trading Signals，將其新增到工作區。
為需要使用其工具的對話啟用該服務。

Web executable，透過 Streamable HTTP。遠端服務在工作區中設定後即可從網頁執行環境執行。

交給代理程式安裝

把這段提示詞貼上到 Claude Code、Codex、Cursor 等代理程式，或 SourceWeft 對話中。代理程式會讀取該伺服器的安裝指南，先向你展示發布者、端點和登入方式，經你同意後再把它新增到你的客戶端。金鑰不會寫進任何命令。

閱讀 https://sourceweft.com/mcp/io.github.roman-rr%2Ftrading-signals/install.md，並按其中說明把這個 MCP 伺服器接入我的客戶端。修改任何設定之前，先告訴我它的發布者、端點和登入方式，等我確認。不要把 API 金鑰寫進命令或我會提交的檔案；問我金鑰應從哪裡取得。

檢視代理程式所依據的安裝指南

其他 MCP 客戶端

把它新增到你客戶端的 mcpServers 設定中。

{
  "mcpServers": {
    "trading-signals": {
      "type": "http",
      "url": "https://signals.x70.ai/mcp"
    }
  }
}

README

Trading Signals

[Trading Signals for AI Agents]

17 triggers. 44 algorithms. 3 AI experts. Every signal backed by causal reasoning and verified against real prices.

Live crypto trading signals for Claude Code, OpenAI Codex, Cursor, Windsurf & 30+ AI agents. Every signal comes with its reasoning — a transmission chain showing exactly which data points led to the trade — and is verified against real prices. Pro adds the trade levels: entry, stop-loss, take-profit, leverage and position size. See Pricing.

Updates, new features and how-tos: follow @x70signals on X.

Install

Claude Code Plugin (recommended)

npx (Claude Code, Codex, Cursor, Windsurf, etc.)

MCP Server (Claude, Cursor, VS Code, Codex)

Sign in from your agent (recommended). Add the URL below and your client opens a browser page where you sign in or create a free account — 14 days of Pro, no card — and click Allow. No key to copy.

Claude Code

bash

claude mcp add --transport http trading-signals https://signals.x70.ai/mcp/connect# then in Claude Code: /mcp → trading-signals → Authenticate

claude mcp add --transport http trading-signals https://signals.x70.ai/mcp/connect# then in Claude Code: /mcp → trading-signals → Authenticate

Claude app (Desktop & claude.ai) — Customize → Connectors → Add custom connector, URL https://signals.x70.ai/mcp/connect.

Cursor (~/.cursor/mcp.json)

Try it without an account: use https://signals.x70.ai/mcp instead — 3 live signals plus the verified hit rate. When you ask for more, your client offers to sign in.

With an API key (Codex, scripts, clients that can't sign in) — get a free key at https://signals.x70.ai/mcp-signup?utm_content=github-readme; the API Key page has copy-paste setup for every client.

bash

claude mcp add --transport http trading-signals https://signals.x70.ai/mcp --header "X-Api-Key: YOUR_KEY"

claude mcp add --transport http trading-signals https://signals.x70.ai/mcp --header "X-Api-Key: YOUR_KEY"

Claude app with a key: https://signals.x70.ai/mcp?apiKey=YOUR_KEY (connectors have no header field).

Codex (~/.codex/config.toml, key in env var SIGNALS_API_KEY)

toml

[mcp_servers.trading-signals]url = "https://signals.x70.ai/mcp"bearer_token_env_var = "SIGNALS_API_KEY"

[mcp_servers.trading-signals]url = "https://signals.x70.ai/mcp"bearer_token_env_var = "SIGNALS_API_KEY"

Tools: get_signals, get_signal, get_signal_history, get_stats (register is deprecated — keys come from the dashboard).

OpenAI Codex CLI

See agents/openai.yaml for agent configuration.

Manual

bash

# Claude Codegit clone https://github.com/roman-rr/trading-skills.gitcp -r trading-skills/trading-signals ~/.claude/skills/trading-signals
# OpenAI Codex CLIcp -r trading-skills/trading-signals ~/.codex/skills/trading-signals

# Claude Codegit clone https://github.com/roman-rr/trading-skills.gitcp -r trading-skills/trading-signals ~/.claude/skills/trading-signals
# OpenAI Codex CLIcp -r trading-skills/trading-signals ~/.codex/skills/trading-signals

Always Watching

17 Triggers. 6 Groups. Every Minute.

The sentinel continuously scans 50+ perpetual markets for anomalies across six orthogonal dimensions -- volume, positioning, price dynamics, cross-asset flows, microstructure, and options-derived signals. A signal fires only when multiple independent dimensions agree, filtering noise from genuine opportunities.

Multi-Expert Consensus

Three specialized AI experts analyze every opportunity independently -- each with a different lens on market structure. Signals require agreement; conflicting views are flagged or filtered. The system continuously learns which analytical approaches perform best per market condition.

Research-Grounded

44 Scientific Methods. 53 Academic Citations.

Every algorithm is grounded in peer-reviewed research from quantitative finance, statistics, and machine learning -- spanning position sizing, anomaly detection, technical analysis, market microstructure, and adaptive risk management.

What You Get

Feature	Description
Live signals	Bullish/bearish with full trade setup -- entry, SL, TP, leverage, position size
Transmission chains	2-4 causal reasoning steps per signal with specific data points
50+ coins	Dynamically selected from top volume, funding, and OI on Hyperliquid perps
Confidence score	0-100 AI conviction level per signal
Auto-verification	Every signal tracked against real prices -- TP/SL monitoring every minute
Paper trading P&L	Real trades on a virtual account with optimal position sizing
Dynamic risk control	Confidence-based leverage caps, volatility-calibrated SL, automatic regime detection
Performance stats	Hit rate, cumulative ROI, profit factor, breakdown by direction/coin/model

Signal Preview

Example rows — not live data. Entry, SL, TP and leverage are returned on Pro; the free plan returns coin, direction, confidence and the reasoning.

API Endpoints

Method	Path	Auth	Description
POST	/api/skill/register	No	Deprecated — returns `signup_required`; get keys at /mcp-signup
GET	/api/skill/signals	API key	List signals (active/verified/all)
GET	/api/skill/signals/:id	API key	Single signal detail
GET	/api/skill/stats	API key	Performance statistics

Example Prompts

Once installed, ask your AI agent:

"Get me the latest crypto trading signals"
"Show me today's highest-confidence BTC and ETH signals"
"What's the 30-day hit rate for these trading signals?"
"Build me a Python script that fetches crypto signals and alerts me on Telegram"
"Show me verified signals from the last week -- what hit TP?"

Paper-trade it first

Want to see how the signals do before you trade them? Let your agent run the paper-trading bot on your machine. It takes each new signal as a $100 paper position at the live price, charges fees on both sides, and logs every trade, wins and losses, to a CSV you own. No real orders, no exchange account; one Python file, standard library only.

The API Key page has a ready prompt for Claude Code, Codex or Cursor, with your key filled in. Our own copy has been running since Oct 7: its live record.

Pricing

	No key	Free key	Pro
Signals	3-signal preview	every live signal — coin, direction, confidence, reasoning	everything in Free + entry, stop-loss, take-profit, leverage, position size
History	—	24 hours	30 days
Rate limit	30 calls/hour per IP	300 calls/hour	effectively unlimited
Price	free	free	$35/month or $300/year

New accounts get a 14-day Pro trial — no card. Nothing is charged unless you subscribe. Pricing & sign-up

Contributing

Issues, feature requests, and PRs are welcome. Please open an issue first to discuss changes.

If this is useful, a ⭐ helps other people find it.

License

Proprietary — the source is readable; a paid license is required for commercial redistribution. See LICENSE.txt.

來源：README.md，提交 3424b52

工具

register

Deprecated: returns a sign-up link instead of creating an API key, ignores its arguments and stores nothing. Don't call it to get access — every other tool works without a key (get_signals returns a 3-signal preview, get_stats is public). To get the full feed, users sign in from the agent at https://signals.x70.ai/mcp/connect (the client opens the browser), or create a key at https://signals.x70.ai/mcp-signup and set it as the X-Api-Key header. Kept only so older clients don't break.

輸入參數nameemailpurposegithubUrl

get_signals

Get Live Trading Signals

List live crypto trading signals that are still open — coin, direction, confidence and the reasoning or market data behind each. Use it to find current trade ideas; use get_signal for one signal by ID (live or resolved), get_signal_history for signals that already closed, and get_stats for aggregate hit rate. No key: a 3-signal preview. Free key: up to 100 signals, with entry, stop-loss, take-profit and leverage null (the response's announcement.upgradeUrl explains Pro). Pro: all levels. If a response contains an announcement, pass its message on to the user — it is about their account (trial ending, plan limits). Returns { signals[], returned, totalMatching, filtersApplied, plan }, sorted by live PnL (best first). Optional filters narrow the list: direction, coins, confidence band, risk/reward, stop distance, open interest, 24h volume, tags, the anomaly that triggered it, days/hours to skip, a BTC 24h-return gate. Rate limits: 30 calls/hour per IP without a key, 300/hour on a free key.

輸入參數coindayslimitapiKeyskipDOWcoinOiMaxcoinOiMindirectioncoinVolMincoinExcludecoinIncludeskipHourUtc另有 10 個

get_signal

Get Signal Detail

Get one signal by its ID — live or already resolved — with its reasoning, live price, PnL and verification outcome (status pending, success or failed; whether take-profit or stop-loss was hit). Use it to follow up a signal from get_signals or get_signal_history, or to see how a past signal ended. Needs an API key (free or Pro); on the free plan entry, stop-loss, take-profit and leverage are null. Returns { success, signal, plan }, or { error: 'Signal not found' } for an unknown ID.

輸入參數apiKeysignalId必填

get_stats

Get Trading Performance Stats

Aggregate performance of signals created in the last N days (default 30) that have since resolved: hit rate, verified count, average leverage and a breakdown by direction; with an API key also average ROI per signal and cumulative ROI. byMonth: hit rate this month, the two months before and all time — the figures shown on signals.x70.ai. Use it to judge recent reliability before acting — it returns no individual signals (use get_signals or get_signal_history for those). Public: works without a key. Past results don't predict future results.

輸入參數daysapiKey

get_signal_history

Get Signal History

List signals that have already resolved (success or failed), newest first, each with its verification outcome. Use it to review past performance signal by signal; for open signals use get_signals, for one signal by ID get_signal, for totals get_stats. Needs an API key. History depth: last 24 hours on the free plan; on Pro the days argument (up to 90). Entry, stop-loss, take-profit and leverage are null on free. Paginate by passing the last signal ID as cursor. Returns { signals[], nextCursor, hasMore, historyDepthDays, plan }.

輸入參數coindayslimitapiKeycursorstatus