What are the runtime requirements for Knowerage?
Node.js 18 or newer is required, and `npx` must be on your `PATH`.
@MTimma
基本情報
Local MCP server that allows your agent to track documentation coverage over code repository.
以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。
{
"mcpServers": {
"knowerage": {
"command": "npx",
"args": [
"@mtimma/knowerage"
],
"env": {
"KNOWERAGE_WORKSPACE_ROOT": "${workspaceFolder}",
"KNOWERAGE_AUTO_FULL_RECONCILE": "true"
}
}
}
}ツール一覧はまだありません
この掲載には表示できるツール一覧がまだありません。サーバーにツールがないという意味ではありません。対応機能はプロジェクトのドキュメントをご確認ください。
上のボタンで一時的なサンドボックスでこのサーバーを実行し、ツール一覧を取得できます。
Knowerage is an MCP server that manages AI analysis coverage of a codebase. It tracks which files and line ranges have been analysed by an AI assistant, storing results as markdown documents and a registry. It is intended for developers using AI coding assistants who want to measure and manage what parts of their project have been covered.
Register the server in your MCP host configuration using npx @mtimma/knowerage and set the KNOWERAGE_WORKSPACE_ROOT environment variable to your project root. Once configured, you talk to your AI assistant in normal sentences—for example, “analyse the logical algorithm workflow in main.java” or “what part of this codebase is not yet analysed?”. The assistant creates or updates analysis documents under knowerage/analysis/ and records coverage in knowerage/registry.json.
Node.js 18 or newer is required, and `npx` must be on your `PATH`.
Set the environment variable `KNOWERAGE_AUTO_FULL_RECONCILE` to a truthy value (`1`, `true`, `yes`, or `on`). The server then watches the `knowerage/` directory and runs a short debounced full reconciliation on filesystem changes.
The file watcher ignores writes to `registry.json`, so saving the registry does not trigger another reconciliation.
All analysis markdown files live under `knowerage/analysis/`, and the coverage registry is stored at `knowerage/registry.json` in the project root.
All file paths are validated against the workspace root; path traversal (`..`) is rejected. Registry writes use atomic file operations for crash safety. No secrets are stored in analysis files or reports.
YC (S26) | Open Computer History | Continuously record your company computer work, map your workflows, help you find work worth automating, and power your agents' context

Handl speaks the Model Context Protocol. Your whole billing operation — projects, invoices, clients, reports, approvals — as tools any AI assistant can call, inside your autonomy rules.
A curated list of references for MLOps

Yuppsy MCP: build and publish real estate websites from your AI assistant. Yuppsy lets your AI assistant create live websites for real estate agents and individual property listings. You don't need a designer or any code
A one-of-a-kind resume builder that keeps your privacy in mind. Completely secure, customizable, portable, open-source and free forever. Try it out today!
コメント