ProdMoh/prodmoh-mcp

prodmoh-mcp

Bring your PRDs & User Stories directly into your IDE — via MCP Prodmoh MCP is a Model Context Protocol (MCP) server that connects your Prodmoh workspace to any MCP-compatible AI client (ChatGPT, Claude Desktop, VS Code agents, custom IDE tools).

ProdMoh MCP Server 🚀

Your PRDs & User Stories, delivered directly into your IDE.

ProdMoh MCP is a Model Context Protocol (MCP) server that connects your ProdMoh workspace to any MCP-compatible AI client (Cursor, VS Code, Claude Desktop, Windsurf).

It acts as your Product Context Layer, delivering PRDs, User Stories, and Acceptance Criteria directly into your coding environment so developers can build, test, and ship with perfect alignment.

⚡ Why ProdMoh MCP?

Today’s development workflow is broken. Developers constantly context-switch between:

• IDE (Code)
• Jira / Linear (Tasks)
• Confluence / Notion (Docs)
• Slack (Clarifications)

ProdMoh MCP fixes this. It makes product requirements instantly accessible inside AI-powered coding workflows.

• No context switching.
• No outdated docs.
• No hallucinations.

🧑‍💻 Developer Workflows & Prompt Library

Maximize your productivity by using these prompt patterns in Cursor, VS Code, or Claude Desktop.

1. 🏗️ Scaffold Implementation Code (Backend)

Stop guessing field names. Generate services that match the spec exactly.

Prompt: "Using get_prd, fetch the 'Checkout Flow' requirements. Generate the Node.js service for this feature. Include validation logic for all edge cases listed in the doc."

2. 🧪 Auto-Generate Unit Tests (TDD)

Achieve 100% coverage of Acceptance Criteria before writing logic.

Prompt: "Fetch the User Story for 'Login'. Generate a Jest test suite that covers every bullet point in the Acceptance Criteria. Ensure you include test cases for the 'Negative Scenarios' (e.g., locked account, timeout)."

3. 🎨 Frontend Component Construction

Build UI components that respect design constraints and accessibility rules.

Prompt: "Read the 'User Profile' User Story. Scaffold a React/Tailwind component for the profile card. Ensure the 'Bio' field enforces the character limit defined in the requirements and includes the specific aria-labels mentioned."

4. 🗄️ Database Schema Modeling

Translate data requirements directly into SQL or ORM definitions.

Prompt: "Reference the 'Order Management' PRD. Generate the Prisma schema model for Orders and OrderItems. Ensure all relationship constraints (1-to-1 vs 1-to-many) match the Data Requirements section."

5. 📜 API Contract Generation

Prevent frontend/backend drift by generating specs from the source of truth.

Prompt: "Look at the 'Search API' requirements. Generate an OpenAPI (Swagger) YAML definition for this endpoint, including all the specific error codes (400, 403, 429) defined in the 'Error Handling' section."

6. 🕵️ Code Review & Validation

Use the AI as a 'QA Buddy' to check your work against the spec.

Prompt: "I have highlighted my current implementation of the calculateTax function. Compare this code against the 'Taxation Rules' PRD. Did I miss any specific state exemptions or rounding rules mentioned in the doc?"

7. 💡 Clarify Requirements Instantly

Don't wait for a Zoom call to get an answer.

Prompt: "@ProdMoh What are the specific password complexity rules defined in the Security PRD? Do we require special characters?"

🛠️ Installation & Setup

You do not need to run a local server. ProdMoh provides a hosted MCP endpoint via Server-Sent Events (SSE).

Prerequisite

1. Log in to your ProdMoh Dashboard.
2. Click "Share with Developers" to generate your secure x-prodmoh-token.

Option 1: Configure Cursor / VS Code (Recommended)

Add this to your mcp.json config file (usually found in ~/.cursor/mcp.json or %APPDATA%/Cursor/User/globalStorage/mcp.json).

{
  "mcpServers": {
    "prodmoh": {
      "type": "sse",
      "url": "[https://api.prodmoh.com/mcp/sse](https://api.prodmoh.com/mcp/sse)",
      "headers": {
        "x-prodmoh-token": "YOUR_TOKEN_HERE"
      }
    }
  }
}

{
  "mcpServers": {
    "prodmoh": {
      "type": "sse",
      "url": "[https://api.prodmoh.com/mcp/sse](https://api.prodmoh.com/mcp/sse)",
      "headers": {
        "x-prodmoh-token": "YOUR_TOKEN_HERE"
      }
    }
  }
}

What Prodmoh MCP Does

Direct PRD Retrieval

Fetch any PRD by:

• Name
• ID
• Feature keyword
• Tags
• Semantic search

User Story Access

Pull complete user stories for features or epics:

• Story text
• Acceptance criteria
• Status & priority

Requirements & Acceptance Criteria Extraction

LLMs use this for:

• Code generation
• Test generation
• API scaffolding
• Validation logic

Documentation Search

Search all Prodmoh workspace docs:

• PRDs
• Stories
• Requirements
• Insights
• Edge cases

Context for Coding, Testing & Refactoring

Zero assumptions.
Only accurate product truth.

How Developers Use Prodmoh MCP for Coding & Testing

Code & tests generated with real product context — without leaving the IDE.

Prodmoh MCP enables a new AI-native developer workflow:

1. Pull PRDs & stories directly inside ChatGPT / IDE

Use Prodmoh MCP to get_prd "Checkout Flow"

Instantly retrieves:

• Requirements
• Acceptance criteria
• Edge cases
• Metrics
• Dependencies

Developers don’t search. They just continue coding.

2. Generate implementation code using the real PRD

Ask: Using this PRD, generate the Node.js service. Include validation, routing, and all edge cases.

LLMs now write code based on:

• Actual requirements
• Real edge cases
• Defined constraints

No hallucination.
No misinterpretation.

3. Auto-generate unit tests from acceptance criteria

Generate Jest unit tests using the acceptance criteria from this PRD or user story.

AI produces:

• Meaningful test cases
• Proper coverage
• Expected edge cases
• Failure conditions

This drastically improves correctness & reduces bugs.

4. Ask clarifying questions instantly

No Slack messages. No waiting.

• What happens if the user cancels during onboarding?
• What error states are defined in the PRD?
• What are the password rules mentioned in the Login PRD?

MCP responds using the authoritative product documentation.

5. Continue coding with perfect alignment

Developers build exactly what product intended.
Testing aligns to real acceptance criteria.
No surprises later.

Available MCP Actions -get_prd

Fetches a complete PRD.

Input:

{ "query": "checkout flow" }

Output includes:

-Product overview

-Problem statement

-Requirements

-Acceptance criteria

-Edge cases

-Technical notes

-Metrics / success criteria

get_userstories

Retrieve user stories for a feature or epic.

Input:

{ "feature": "Login" }

Output:

-Story ID

-Story text

-Acceptance criteria

-Priority & status

License

MIT License — free to use, modify, and integrate commercially.

Support the Project

If Prodmoh MCP improves your workflow:

⭐ Star the repo

Share it with your team

Contribute improvements

Build IDE plugins using it