Konseki
БесплатноНе проверенEnables AI agents to query historical market context data from the Konseki API, returning raw JSON for analysis.
Описание
Enables AI agents to query historical market context data from the Konseki API, returning raw JSON for analysis.
README
Official Model Context Protocol server for Konseki.
Konseki is a pre-computed historical market context API for global equities, built for AI trading and quants. It matches current market conditions against historical analogs and returns structured pattern-match data, outcome distributions, and match-quality scores, grounding an AI agent's reasoning in evidence rather than generic commentary.
Konseki MCP is a direct wrapper around the Konseki public API. It lets AI agents and AI trading tools call Konseki endpoints through MCP tools while preserving the raw API response JSON for the user or downstream application to interpret.
Published package: @konseki/mcp
Requirements
- Node.js 20 or newer.
- A Konseki API key.
Get an API key from konseki.io.
Design Principles
- Direct API wrapper: the MCP server fetches Konseki API responses and returns them without interpretation.
- User-owned interpretation: users, builders, and downstream AI clients decide how to analyze or summarize the returned JSON.
- Public API only: the server uses
X-API-Keyagainst documented Konseki endpoints. - No internal credentials: non-public service or operational credentials are never required for this package.
Configuration
The server reads configuration from environment variables:
KONSEKI_API_KEY=ks_live_your_api_key
Do not commit real API keys.
Intended Architecture
AI client
|
v
Konseki MCP server
|
v
Konseki public API
|
v
Raw historical market context JSON
The MCP server should not bypass Konseki public API behavior. It should behave like any other public API client.
Tools
get_konseki_metadata
Fetches raw JSON from GET /v1/metadata?country={country}.
Input:
{
"country": "US"
}
list_konseki_countries
Fetches raw JSON from GET /v1/countries.
Input: none.
list_konseki_symbols
Fetches raw JSON from GET /v1/symbols?country={country}.
Input:
{
"country": "US"
}
get_konseki_analysis
Fetches raw JSON from GET /v1/analysis/{symbol}-{exchange}?country={country}&lookback={lookback}.
Input:
{
"country": "US",
"symbol": "AAPL",
"exchange": "NASDAQ",
"lookback": 15
}
Supported lookback values: 5, 10, 15, 20, 25, 30, 40, 50.
Response Compression
The server requests gzip-compressed API responses and decompresses them locally before returning JSON to the MCP client. This is handled automatically; users do not need to configure compression.
Installation
Use the published npm package through an MCP client with npx:
{
"mcpServers": {
"konseki": {
"command": "npx",
"args": ["-y", "@konseki/mcp"],
"env": {
"KONSEKI_API_KEY": "ks_live_your_api_key"
}
}
}
}
This is the recommended configuration for users who want the official released package from npm.
Local Development
For development from this checkout, install dependencies, build the package, and configure your MCP client to run the built server:
npm install
npm run build
{
"mcpServers": {
"konseki": {
"command": "node",
"args": ["/absolute/path/to/konseki-mcp/dist/index.js"],
"env": {
"KONSEKI_API_KEY": "ks_live_your_api_key"
}
}
}
}
Development
Install dependencies:
npm install
Run verification:
npm run typecheck
npm test
npm run build
Security
- Never commit real API keys.
- Never log raw API keys.
- Never include user credentials in test snapshots or examples.
- Use fake keys in documentation and tests.
Установка Konseki
У этого сервера нет опубликованного пакета — он собирается из исходников. Открой репозиторий и следуй инструкции в README.
▸ github.com/konseki-official/konseki-mcpFAQ
Konseki MCP бесплатный?
Да, Konseki MCP бесплатный — установка в пару кликов через Unyly без оплаты.
Нужен ли API-ключ для Konseki?
Нет, Konseki работает без API-ключей и переменных окружения.
Konseki — hosted или self-hosted?
Доступен hosted-вариант: Unyly запускает сервер в облаке, локальная установка не обязательна.
Как установить Konseki в Claude Desktop, Claude Code или Cursor?
Открой Konseki на unyly.org, выбери вкладку своего клиента (Claude Desktop, Claude Code, Cursor) и нажми Install — конфиг сгенерируется автоматически, без правки JSON.
Похожие MCP
Fetch
Web content fetching and conversion for efficient LLM usage.
AWS KB Retrieval
Retrieval from AWS Knowledge Base using Bedrock Agent Runtime.
автор: modelcontextprotocolSpring AI MCP Server
Provides auto-configuration for setting up an MCP server in Spring Boot applications.
llm-analysis-assistant
A very streamlined mcp client that supports calling and monitoring stdio/sse/streamableHttp, and can also view request responses through the /logs page. It also
автор: xuzexin-hzCompare Konseki with
Не уверен что выбрать?
Найди свой стек за 60 секунд
Автор?
Embed-бейдж для README
Похожее
Все в категории ai
