Catalyst Edge Mcp

com.catalystedgescanner由 catalystedgescanner.com 發布官方發布者v1.0.2更新於 Oct 24, 2026

Audited SEC catalyst signals for agents — ranked filing intel with a public track record.

已驗證Streamable HTTP可網頁執行金融資料與分析

概覽

AI 產生的概覽

託管型 MCP 伺服器，讓助理取得依評分排序的美股 SEC 申報催化劑選股、個股訊號拆解、論點、價格預測與選擇權參考。

可以做什麼
  • 取得當日評分最高的催化劑選股，可依信心等級或產業板塊篩選，並查看歷史選股清單的命中率與超額報酬
  • 檢視單一個股的完整收斂拆解，包含每個非零訊號層及其分數、總分與信心等級
  • 閱讀個股的淺白論點，涵蓋量化解讀、催化劑、風險與看空情境
  • 取得 Kronos 價格預測，以收高機率與 p10–p90 報酬區間呈現
  • 查看各產業板塊的彙總方向傾向，以及最近到期日、價平買權／賣權價格與跨式成本等即時選擇權資訊
適用情境
適合助理回答美股研究、以 SEC 申報作為交易催化劑的問題，或協助挑選選擇權策略時使用；若想直接接上具公開歷史紀錄的託管訊號來源，而不自行建置訊號流程，也可考慮。
執行需求
遠端 streamable HTTP 端點，無需安裝。免費評估層不需金鑰，回傳前 3 筆選股。付費 intelligence 工具需在 Authorization 標頭提供 Bearer 權杖，可透過服務商定價頁或電子郵件取得。README 另提及自架 stdio Python 方式。
安裝前請注意
付費層需要 Bearer 權杖，屬應妥善保管的憑證。輸出為研究內容而非投資建議，歷史績效由服務商自行評定。行情資料有延遲，來源包括 SEC EDGAR、政府開放資料及第三方。工具皆為唯讀查詢，不涉及交易或付款動作。

安裝

在 SourceWeft 中

  1. 開啟 儀表板中的 Catalyst Edge Mcp，將其新增到工作區。
  2. 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/com.catalystedgescanner%2Fcatalyst-edge-mcp/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": {
    "catalyst-edge": {
      "type": "http",
      "url": "https://catalystedgescanner.com/mcp/"
    }
  }
}

README

Catalyst Edge MCP Server

Publicly graded SEC catalyst intelligence for AI agents, over the Model Context Protocol. Every US-market SEC filing with market-moving potential is fetched from EDGAR, scored by 500+ data engines, and ranked before the market opens — and unlike most signal products, the track record is public: every past call is published with its outcome at catalystedgescanner.com/receipts.

Use the hosted server (no install)

A hosted instance is live. Point any MCP-capable client at it, or POST JSON-RPC directly:

bash
curl -s https://catalystedgescanner.com/mcp/ \  -H "Content-Type: application/json" \  -d '{"jsonrpc":"2.0","id":1,"method":"tools/list"}'

Call a tool (no key needed for the free tier):

bash
curl -s https://catalystedgescanner.com/mcp/ \  -H "Content-Type: application/json" \  -d '{"jsonrpc":"2.0","id":2,"method":"tools/call","params":{"name":"get_convergence_picks","arguments":{"limit":3}}}'

Paid tiers use a Bearer token: Authorization: Bearer <your-key>.

Tools

ToolTierReturns
get_convergence_picksfreetoday's scored catalyst picks (free: top 3)
get_track_recordfreehistorical hit-rate / alpha of published picks
get_ticker_signalintelligencefull per-ticker signal-layer breakdown
get_thesisintelligenceplain-language thesis (catalysts, risks, bear case)
get_sector_leanintelligencedirectional sector lean

free — evaluation tier, no key required. intelligence — all tools, full depth; keys via catalystedgescanner.com/pricing or [email protected].

Claude Desktop / Cursor config (stdio, self-hosted)

json
{  "mcpServers": {    "catalyst-edge": {      "command": "python3",      "args": ["/path/to/catalyst_mcp.py"],      "env": { "CATALYST_MCP_KEY": "optional-key" }    }  }}

Stdlib Python only — no dependencies. HTTP mode: CATALYST_MCP_HTTP=1 python3 catalyst_mcp.py (binds 127.0.0.1:8848; front with TLS for public use — see catalyst-mcp.service). Key→tier mapping lives in mcp_keys.json (see mcp_keys.json.example).

Verify

bash
python3 smoke_test.py   # exit 0 = handshake + all tools OK

Data & disclaimers

Sources: SEC EDGAR (filings, XBRL, insider activity), US government open data, delayed market prices. Nothing here is financial advice; signals are research with a published, self-graded track record — hits and misses both, each measured against the next session's open. Machine-readable site map: catalystedgescanner.com/llms.txt · OpenAPI: /openapi.json.

MIT licensed.

來源：README.md，提交 5d97ef7

工具

7

get_convergence_picks

Today's top scored catalyst picks from the Catalyst Edge convergence model. Optional filters: conviction, sector. Free tier returns the top 3.

輸入參數limitsectorconviction

get_ticker_signal

Full convergence breakdown for one ticker — every non-zero signal layer and its points, plus the score and conviction.

輸入參數ticker必填

get_thesis

Plain-language thesis for a ticker — quantitative read, catalysts, risks, bear case. Covers the top convergence picks.

輸入參數ticker必填

get_price_forecast

Kronos price forecast for a ticker (or the whole scored board), as a probability: p_up is the share of sampled price paths closing higher over the horizon, with a p10-p90 return band. Price history only — independent of the catalyst signal layers.

輸入參數limitticker

get_sector_lean

Orphan-aggregated sector lean — the directional signal each sector is showing, from ~480 aggregated data spokes.

無需輸入參數。

get_options_context

Live options context for a ticker via Tradier — nearest expiration, ATM call/put prices, straddle cost. Use to choose an options strategy.

輸入參數ticker必填

get_track_record

Historical hit-rate and alpha of Catalyst Edge pick lists — the evidence an agent uses to weight the signal. Optional: list_name filter.

輸入參數list_name

版本歷史

1
  1. v1.0.2最新Sep 16, 2026

評分

已滿足適用檢查項的 100%

該評分衡量此收錄條目的完整程度，依據註冊中心條目、端點與公開來源計算得出。它不是對伺服器本身的評價，也不表示伺服器安全或目前可用：可連通性是帶日期的檢查結果，評分也不包含安裝量或使用者評價。

  • 近 30 天內可連通必需25 分
  • 已宣告認證方式必需10 分
  • 已列出工具必需15 分
  • 描述必需15 分
  • 圖示10 分
  • 文件15 分
  • 授權條款10 分