Procore MCP Server

@mohllal

リポジトリ

基本情報

カテゴリ

その他

ライセンス

MIT license

トランスポート

stdio

作者

mohllal

Procore MCP Server について

MCP server for Procore APIs

#mcp-server

Procore MCP Server の設定

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": {
    "procore-mcp-server": {
      "command": "uv",
      "args": [
        "sync"
      ]
    }
  }
}

Procore MCP Server のツール

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。

ツールは README から自動的に抽出されます。メンテナーは ## Tools という見出しの下に記載することで、このタブに反映できます。

概要

What is Procore MCP Server?

The Procore MCP Server is a Model Context Protocol (MCP) server that connects LLMs to Procore APIs, enabling AI assistants to query and retrieve data from a Procore account. It is designed for developers and construction professionals who want to use natural language to interact with Procore projects, companies, and submittals.

How to use Procore MCP Server?

Install the server with uv sync, then create a .env file with your Procore API credentials (client ID, client secret, and redirect URI). Run python init_token.py to complete OAuth authentication and save a token to procore_token.json. Configure the server in Claude or VSCode by adding a JSON block that invokes uv run procore-mcp-server with the correct directory path.

Key features of Procore MCP Server

Supports setting company and project context via procore_set_context
Lists all accessible companies with procore_list_companies
Retrieves company and project details
Lists projects for a given company
Lists submittals with optional filters
Provides three prompts for generating summaries

Use cases of Procore MCP Server

Quickly view project details without navigating the Procore web interface
Summarize company information from an LLM‑powered chat interface
List and filter submittals across projects for status updates
Integrate Procore data into a custom AI assistant or IDE

よくある質問

What prerequisites are needed to run the Procore MCP Server?

Python 3.13 or later, a Procore account with API access, and API credentials (client ID and client secret) are required.

How does authentication work?

The server uses OAuth 2.0. Run `python init_token.py` to open an authorization URL, log in to Procore, paste the provided authorization code, and the script saves the token to `procore_token.json`. Use `urn:ietf:wg:oauth:2.0:oob` as the redirect URI for redirectless OAuth.

What transport does the Procore MCP Server use?

The server uses the stdio transport. Both Claude and VSCode integrations are configured with `"type": "stdio"` and `"command": "uv"` with the appropriate arguments.

Where is authentication data stored?

The OAuth token is saved locally in a file named `procore_token.json` in the server’s root directory. The server automatically reads this file for API requests.

What tools does the Procore MCP Server expose?

It provides six tools: `procore_set_context`, `procore_list_companies`, `procore_get_company_details`, `procore_list_projects_for_company`, `procore_get_project_details`, and `procore_list_submittals`.

「その他」の他のコンテンツ

Hindsight

vectorize-io

Hindsight: Agent Memory That Learns

50.4K

Yuppsy

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

DeepMark

DeepMark helps teachers deliver rapid, consistent marking with meaningful feedback for every student — in a fraction of the time. What once took a week, now takes one free period.