What are the runtime requirements for Specifai MCP Server?
Node.js version 16.0.0 or higher is required. Bun 1.0.0 or higher is optional if using the Bun runtime.
Specifai MCP Server
@presidio-oss
基本信息
关于 Specifai MCP Server
MCP server to provide Specifai project requirements, task, etc. to any AI tools supporting MCP such as Hai Build, Cursor.
Specifai MCP Server 配置
使用下面的配置,将此服务器添加到你的 MCP 客户端。
{
"mcpServers": {
"specifai-mcp-server": {
"command": "npx",
"args": [
"--yes",
"@presidio-dev/specifai-mcp-server@latest"
]
}
}
}Specifai MCP Server 工具
13Get Business Requirement Documents for this project, returns ID, Title, Description(not included by default)
Get Product Requirement Documents for this project, returns ID, Title, Description(not included by default), Ado Id (When PRD/US/Task is linked to ADO), Jira Id (When PRD/US/Task is linked to JIRA), PMO Id(when available), Ado Issue Type(when available), Jira Issue Type(when available), and LinkedBRDIds (when available)
Get Non-Functional Requirement Documents for this project, returns ID, Title, Description(not included by default)
Get User Interface Requirement Documents for this project
Get Business Process Documents for this project
Get Test Case Documents for this project
Get User Stories for a particular PRD
Get Tasks for a particular User Story
Get list of Tasks for a particular User Story in a particular PRD
Update the project path and reload the solution. use this tool only when we not automatically able to infer the project path or asked by the user or us., we will try to auto infer it from the environment first.
Retrieves complete information about a specific task when you know its exact ID. This tool provides direct access to a single task's details without having to retrieve and filter through all tasks., returns full task object, ID, Title, Description, Ado Id (When Task is linked to ADO), Jira Id (When Task is linked to JIRA), PMO Id(when available), Ado Issue Type(when available), Jira Issue Type(when available)
List all the tasks available across all PRDs and User Stories, without task description, only return ID, Title, Ado Id (When Task is linked to ADO), Jira Id (When Task is linked to JIRA), PMO Id(when available), Ado Issue Type(when available), Jira Issue Type(when available)
Search documents by text content or Pmo ID across all documents, returns document ID, Title, Description, Type, Ado Id (When PRD/US/Task is linked to ADO), Jira Id (When PRD/US/Task is linked to JIRA), PMO Id(when available), Ado Issue Type(when available), Jira Issue Type(when available)
概览
What is Specifai MCP Server?
Specifai MCP Server is a Model Context Protocol server that integrates with the Specifai project to allow any MCP-compatible AI tool to read and interact with specification documents generated by Specifai. It is tool-agnostic and currently exposes tools for reading Business Requirement Documents, Product Requirement Documents, Non-Functional Requirements, User Interface Requirements, Business Process Documents, Test Case Documents, User Stories, Tasks, and more.
How to use Specifai MCP Server?
Install using npx --yes @presidio-dev/specifai-mcp-server@latest and configure your MCP client with the command and arguments. Optionally place a .specifai-path file in your project root containing the absolute path to your Specifai project directory. The server provides tools such as get-brds, get-prds, get-nfrs, get-uirs, get-bpds, get-tcs, get-user-stories, get-tasks, get-task, set-project-path, get-task-by-id, list-all-tasks, and search.
Key features of Specifai MCP Server
- Read all specification documents generated by Specifai
- Tool‑agnostic – works with any MCP‑compatible AI tool
- Full‑text search across all specification documents
- Supports multiple IDEs: VS Code, Cursor, Windsurf, Zed, Hai Build, Amazon Q Developer
- Optional
.specifai-pathfile for persistent project directory configuration - Simple JSON configuration via
mcpServersblock
Use cases of Specifai MCP Server
- AI‑assisted code generation based on Business Requirement Documents (BRDs)
- Automated test case creation from Test Case Documents (TCs)
- User story and task breakdown for agile development workflows
- Querying non‑functional requirements during architecture discussions
- Searching across all specification documents for relevant information
常见问题
How does Specifai MCP Server know where my project specifications are?
Create a `.specifai-path` text file in your project root containing the absolute path to the directory where Specifai stored the specification documents. Optionally, you can use the `set-project-path` tool at runtime.
Which IDEs and extensions are supported for Specifai MCP Server?
Hai Build, Amazon Q Developer, VS Code (Copilot), Cursor, Windsurf, and Zed are all supported with documented configuration steps.
Is Specifai MCP Server stable for production use?
No, the server is currently experimental. Its functionality and available tools are subject to change as development continues.
What transport protocol does Specifai MCP Server use?
The server uses the stdio transport, as shown in the standard configuration examples (e.g., `"type": "stdio"` for VS Code).
开发工具 分类下的更多 MCP 服务器
SSH MCP Server
hypnosisSSH MCP server for AI agents: remote commands, file transfer, log search and server audits through OpenSSH.
1
FARPY
mangomunchrVerified distributed compute for humans, applications, and agents. Submit Blender, Octane, and video workloads, route them across qualified GPUs, track asynchronous execution, download finished artifacts, and verify FARP
1
offprompt
offpromptMCP server that lets Claude Code, Cursor or Codex ask you for API keys on a local page and write them to .env. The agent only ever sees key names.
2
ArtiBot Cloud MCP - Zhijiangyun 智匠云
bofobArtiBot Cloud MCP (Zhijiangyun 智匠云) — cloud MCP server for robot manipulation data & GPU workloads. 15 tools: robot fleet management, data collection & video recording, post-processing pipelines, and benchmark evaluation
1
Casefile
azimov777One case file per task — shared by every session, agent and person that touches it. Self-hosted MCP server and live board. MIT.
3
评论