mcpField
S
Specifai MCP Server

Specifai MCP Server

Local
github.com/presidio-ossDeveloper Tools13Updated 65w ago

About Specifai MCP Server

Specifai MCP Server gives MCP-compatible AI tools access to project documents generated by Specifai. It supports document retrieval, task queries, and full-text search; the project is experimental.

Key features

  1. 01Read business and product requirements
  2. 02Read non-functional and UI requirements
  3. 03Read process documents and test cases
  4. 04Query user stories and tasks
  5. 05Search across documents

Use cases

  1. 01Review project requirements in an IDE
  2. 02Find tasks associated with a user story
  3. 03Search Specifai project documents
  4. 04Look up task details by ID

MCP support

The server exposes Specifai document, search, and task tools to MCP-compatible clients, including tools such as Hai Build and Cursor.

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

mcp.json · Universal config
{
"mcpServers": {
"specifai": {
"command": "npx",
"args": [
"-y",
"@presidio-dev/specifai-mcp-server"
]
}
}
}
Run commandnpx -y @presidio-dev/specifai-mcp-server

Local servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.

Packages
npm@presidio-dev/specifai-mcp-server

Tools

13

Saved tool list · updated 3w ago

ParameterTypeRequiredDescription
cwdstringYesAbsolute path where the tool is called from and containing the `.specifai-path` file to auto-infer the specifai project path. This path will be current working directory (cwd) from where the tool is called.
includeDescriptionbooleanNoInclude description in the output, default is false
limitnumberNoLimit the number of documents to return, default is 10
offsetnumberNoOffset the number of documents to return, default is 0

Details

From the project README

Specifai MCP Server

License: MIT npm version Install in VS Code (npx) Install in VS Code Insiders (npx) Install in Cursor (npx)

Hai Build Amazon Q VS Code Cursor Windsurf Zed

A Model Context Protocol (MCP) server for Specifai project integration and automation with any MCP-compatible AI tool. This server is designed to be tool-agnostic, meaning it can be used with any tool that supports the MCP protocol. This server currently exposes tools to read all documents generated by the Specifai project.

Warning

This server is currently experimental. The functionality and available tools are subject to change and expansion as we continue to develop and improve the server.

Table of Contents

Requirements

  • Node.js >= 16.0.0
  • Bun >= 1.0.0 (if using Bun runtime)
  • Hai Build, Cursor, Windsurf, Claude Desktop or any MCP Client

Installation

# Latest version
npx --yes @presidio-dev/specifai-mcp-server@latest

# Specific version
npx --yes @presidio-dev/[email protected]

We recommend npx to install the server, but you can use any node package manager of your preference such as yarn, pnpm, bun, etc.

Configuration

with npx with latest version:

{
  "specifai": {
    "command": "npx",
    "args": ["--yes", "@presidio-dev/specifai-mcp-server@latest"],
    "disabled": false,
    "autoApprove": []
  }
}

with npx with specific version:

{
  "specifai": {
    "command": "npx",
    "args": ["--yes", "@presidio-dev/[email protected]"],
    "disabled": false,
    "autoApprove": []
  }
}
Preparing your project

This is completely optional, but it's recommended to use it to avoid having to specify the project directory path every time you access the server. For AI IDE / Extension (Hai Build), it's recommended to use a .specifai-path file to specify the project directory path.

Make sure your project root directory contains a .specifai-path file. It's how the Specifai MCP server knows where to find the specification documents generated by Specifai.

The file is a plain text file containing the absolute path to the project directory where the specification documents for a project are stored.

For example, if your project directory is located at /path/to/project, the .specifai-path file should contain the following line:

/path/to/project

Specifai MCP integration with popular IDE and extension

See the setup instructions for each

Install in Hai Build

Add the following to your hai_mcp_settings.json file. To open this file from Hai Build, click the "MCP Servers" icon, select the "Installed" tab, and then click "Configure MCP Servers".

See the Hai Build MCP documentation for more info.

{
  "mcpServers": {
    "specifai": {
      "command": "npx",
      "args": ["-y", "@presidio-dev/specifai-mcp-server@latest"]
    }
  }
}
Install in Amazon Q Developer

Add the following to your Amazon Q Developer configuration file. See MCP configuration for Q Developer in the IDE for more details.

The configuration file can be stored globally at ~/.aws/amazonq/mcp.json to be available across all your projects, or locally within your project at .amazonq/mcp.json.

{
  "mcpServers": {
    "specifai": {
      "command": "npx",
      "args": ["-y", "@presidio-dev/specifai-mcp-server@latest"]
    }
  }
}
Install in VS Code (Copilot)

Install in VS Code (npx) Install in VS Code Insiders (npx)

First, enable MCP support in VS Code by opening Settings (Ctrl+,), searching for mcp.enabled, and checking the box.

Then, add the following configuration to your user or workspace settings.json file. See the VS Code MCP documentation for more info.

"mcp": {
  "servers": {
    "specifai": {
      "type": "stdio",
      "command": "npx",
      "args": ["-y", "@presidio-dev/specifai-mcp-server@latest"]
    }
  }
}
Install in Cursor

The easiest way to install is with the one-click installation button below.

Install in Cursor (npx)

Alternatively, you can manually configure the server by adding the following to your mcp.json file. This file can be located globally at ~/.cursor/mcp.json or within a specific project at .cursor/mcp.json. See the Cursor MCP documentation for more information.

{
  "mcpServers": {
    "specifai": {
      "command": "npx",
      "args": ["--yes", "@presidio-dev/specifai-mcp-server@latest"]
    }
  }
}
Install in Windsurf

Add the following to your ~/.codeium/windsurf/mcp_config.json file. See the Windsurf MCP documentation for more information.

{
  "mcpServers": {
    "specifai": {
      "command": "npx",
      "args": ["-y", "@presidio-dev/specifai-mcp-server@latest"]
    }
  }
}
Install in Zed

You can add the Specifai MCP server in Zed by editing your settings.json file (accessible via the zed: settings action) or by using the Agent Panel's configuration UI (agent: open configuration). See the Zed MCP documentation for more information.

Add the following to your settings.json:

{
  "context_servers": {
    "specifai": {
      "command": {
        "path": "npx",
        "args": ["-y", "@presidio-dev/specifai-mcp-server@latest"]
      }
    }
  }
}
Available Tools

The server provides several tools for interacting with your specification documents:

Tool NameDescription
get-brdsGet Business Requirement Documents
get-prdsGet Product Requirement Documents
get-nfrsGet Non-Functional Requirements
get-uirsGet User Interface Requirements
get-bpdsGet Business Process Documents
get-tcsGet Test Case Documents
get-user-storiesGet User Stories for a specific PRD
get-tasksGet Tasks for a specific User Story
get-taskGet details of a specific Task
set-project-pathSet or change the project directory path
get-task-by-idGet details of a specific Task by ID
list-all-tasksList all available tasks
searchFull text search across all documents

Contributing

We welcome contributions to the Specifai MCP Server! Please see our Contributing Guide for more information on how to get started.

Development Setup

For detailed instructions on setting up your development environment, please refer to our Development Setup Guide.

Architecture

To understand the project architecture, please see our Architecture Guide.

Security

For information about our security policy and how to report security vulnerabilities, please see our Security Policy.

License

This project is licensed under the MIT License - see the LICENSE file for details.

Acknowledgements

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.