SignalScope
Provides Cursor AI agents with direct access to stock breakout signals, including trending tickers, AI-generated reports, portfolio tracking, and watchlist management, all aggregated and scored from multiple sources.
README
SignalScope — Cursor Plugin
AI-powered stock breakout signal detection inside Cursor IDE
SignalScope gives Cursor AI agents direct access to stock breakout signals via MCP tools. Query trending tickers, deep-dive with AI-generated reports and trade setups, track your portfolio and watchlist — all without leaving the IDE. Signals are aggregated from social media, SEC filings, congressional trades, options flow, and volume spikes, then scored and filtered for pump-and-dump patterns.
What is SignalScope?
SignalScope aggregates signals from social media, SEC filings, congressional trades, options flow, volume spikes, and Polymarket. It scores candidates with AI, filters pump-and-dump patterns, and surfaces validated tickers with opportunityScore (early-mover rank) and aiScore (evidence confidence).
Installation
Install from the Cursor Marketplace by searching for SignalScope.
After installation, configure authentication (pick one):
Option A — API key (subscription required):
- Open Cursor Settings → Features → Model Context Protocol
- Find the
signalscopeserver and click the gear icon - Set
SIGNALSCOPE_API_KEYto your key
To get your API key:
- Create an account at signalscopes.com
- Go to your Profile page
- Click Generate API Key and copy it (shown only once)
Option B — x402 pay-per-call (no subscription needed):
- Set
SIGNALSCOPE_WALLET_PRIVATE_KEYto an Ethereum wallet private key with USDC on Base - Each API call deducts a small USDC amount automatically (e.g. $0.005–$0.05 per call)
- No account or registration required
Note:
search_tickers,get_methodology,list_scans,get_scan,get_signals, andget_platform_statsare public (no key). Monetized ticker endpoints need an API key (subscriber) or x402.generate_reportcannot be generated with an API key — use x402 ($0.05) or sign in on the site (Pro to generate new reports; cached reports may be returned for subscribers).
Manual Installation (without Marketplace)
Add this to your Cursor MCP settings (.cursor/mcp.json or via Settings → MCP):
{
"mcpServers": {
"signalscope": {
"command": "npx",
"args": ["-y", "signalscope-mcp"],
"env": {
"SIGNALSCOPE_API_KEY": "sk_sig_your_key_here"
}
}
}
}
For x402 pay-per-call, replace SIGNALSCOPE_API_KEY with:
"SIGNALSCOPE_WALLET_PRIVATE_KEY": "your_ethereum_private_key"
Available Tools
Discovery (free — no API key)
| Tool | Description |
|---|---|
search_tickers |
Search by symbol or partial name — up to 8 results with aiScore, opportunityScore, stage, and price |
get_methodology |
Platform methodology and scoring pipeline details |
Signal Data
| Tool | Description |
|---|---|
get_trending |
Trending breakout tickers with filtering by stage, trend, market cap, sector, source — x402: $0.01 |
get_ticker |
Latest validated data + raw signals for a symbol — x402: $0.005 |
get_ticker_history |
Historical scan appearances (full list, no pagination) — x402: $0.005 |
get_ticker_performance |
Price performance — 1d, 3d, 7d, 30d returns since signal detection — x402: $0.005 |
get_ticker_related |
Scan co-occurrence candidates ranked by Pearson price correlation — x402: $0.005 |
get_ticker_network |
Price-correlation graph (minCorrelation 0–1); nodes seeded by co-occurrence or trending — x402: $0.01 |
generate_report |
AI report + trade setup — x402 $0.05 or browser session (not via x-api-key) |
list_scans |
List recent harvest scans (public) |
get_scan |
One scan with validated tickers (public) |
get_signals |
Raw signals for a scan (public) |
Portfolio
| Tool | Description |
|---|---|
list_portfolio |
All positions — open and closed |
add_position |
Open a new position |
update_position |
Update or close a position |
delete_position |
Delete a position |
get_portfolio_performance |
Platform-wide performance stats — win rate, avg return by AI score range and weekly cohort |
Watchlist
| Tool | Description |
|---|---|
list_watchlist |
Watchlist enriched with ticker data and performance |
add_to_watchlist |
Add a symbol |
remove_from_watchlist |
Remove a symbol |
Utility
| Tool | Description |
|---|---|
get_prices |
Current prices for up to 50 symbols (5-min cache); auth optional |
get_platform_stats |
Scan count, signal sum, distinct ticker count (public, rate-limited) |
Example Prompts
Once installed, try these prompts in Cursor Chat:
Morning scan:
"Show me the top Consensus-stage breakout signals from SignalScope right now, rising trend only, hiding any P&D flagged ones."
Deep dive:
"Get the full SignalScope report for NVDA including the AI trade setup."
Portfolio check:
"List my SignalScope portfolio and get current prices for all open positions."
Screen by sector:
"Find micro-cap technology tickers with 3+ appearances in SignalScope sorted by AI score."
Correlation research:
"What tickers co-occur most with AMD in SignalScope signals?"
Track platform accuracy:
"Show me the 7-day performance stats from SignalScope — win rate and average return."
Links
- signalscopes.com — Platform
- API Documentation — Full API reference
- Methodology — How signals are scored
- Changelog — What's new
License
MIT
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。