Skip to main content
Glama
prisma-cloud-docs-mcp-server

prisma-cloud-docs-mcp-server

by clarkemn

Prisma Cloud Docs MCP Server

A Model Context Protocol (MCP) server that provides search access to Prisma Cloud documentation. This server allows Claude and other MCP-compatible clients to search through Prisma Cloud's official documentation and API references.

Note: This server has been migrated to HTTP transport and container deployment for improved scalability and performance. The server now runs in HTTP mode when deployed via Smithery.

Features

  • Search across Prisma Cloud documentation

  • Search Prisma Cloud API documentation

  • Caching system for improved performance

  • Real-time indexing of documentation sites

Related MCP server: s2-streamstore

Installation

No installation needed! Just use uvx in your Claude Desktop configuration.

Installing via Smithery

To install prisma-cloud-docs-mcp-server for Claude Desktop automatically via Smithery:

npx -y @smithery/cli install @clarkemn/prisma-cloud-docs-mcp-server --client claude

Option 2: Development Installation

Prerequisites

  • Python 3.12 or higher

  • uv package manager

Install uv

curl -LsSf https://astral.sh/uv/install.sh | sh

Clone and Setup

git clone https://github.com/clarkemn/prisma-cloud-docs-mcp-server.git
cd prisma-cloud-docs-mcp-server
uv sync

Usage

With Claude Desktop

Add this server to your Claude Desktop configuration file:

Location: ~/Library/Application Support/Claude/claude_desktop_config.json (macOS)

{
  "mcpServers": {
    "Prisma Cloud Docs": {
      "command": "uvx",
      "args": ["prisma-cloud-docs-mcp-server@latest"],
      "env": {},
      "transport": "stdio"
    }
  }
}

Option 2: Local Development

{
  "mcpServers": {
    "Prisma Cloud Docs": {
      "command": "uv",
      "args": ["run", "python", "server.py"],
      "cwd": "/path/to/prisma-cloud-docs-mcp-server",
      "env": {},
      "transport": "stdio"
    }
  }
}

Replace /path/to/prisma-cloud-docs-mcp-server with the actual path to where you cloned this repository.

Manual Testing

You can test the server manually:

echo '{"jsonrpc": "2.0", "id": 1, "method": "initialize", "params": {"protocolVersion": "2024-11-05", "capabilities": {}, "clientInfo": {"name": "test", "version": "1.0"}}}' | uv run python server.py

Available Tools

The server provides these MCP tools:

  • index_prisma_docs(max_pages: int = 50) - Index Prisma Cloud documentation (call this first)

  • index_prisma_api_docs(max_pages: int = 50) - Index Prisma Cloud API documentation

  • search_prisma_docs(query: str) - Search Prisma Cloud documentation

  • search_prisma_api_docs(query: str) - Search Prisma Cloud API documentation

  • search_all_docs(query: str) - Search across all indexed documentation

  • get_index_status() - Check indexing status and cache statistics

Development

Running the server

HTTP mode (Production/Smithery):

uv run python -m src.main

STDIO mode (Local development):

uv run python server.py

Container mode:

docker build -t prisma-docs-server .
docker run -p 8081:8081 -e PORT=8081 prisma-docs-server

Installing dependencies

uv sync

Project structure

prisma-cloud-docs-mcp-server/
├── src/
│   ├── main.py           # HTTP MCP server implementation  
│   └── middleware.py     # Configuration middleware for Smithery
├── server.py             # Legacy STDIO server (for local development)
├── pyproject.toml        # Project configuration
├── uv.lock              # Dependency lock file
├── Dockerfile           # Container deployment
├── smithery.yaml        # Smithery container configuration
└── README.md            # This file

License

MIT License - see LICENSE file for details.

Contributing

  1. Fork the repository

  2. Create a feature branch

  3. Make your changes

  4. Test with Claude Desktop

  5. Submit a pull request

Troubleshooting

Server not starting in Claude Desktop

  1. Ensure uv is installed and in your PATH

  2. Verify the path to the project directory is correct

  3. Check Claude Desktop logs for specific error messages

Missing dependencies

Run uv sync to ensure all dependencies are installed.

Documentation not found

The server needs to index documentation first. Use the index_prisma_docs or index_prisma_api_docs tools before searching.

Available Tools

6 tools
get_index_statusA

Check how many documents are currently cached.

Inputs: no named parameters. Returns: result

index_prisma_api_docsA

Index Prisma Cloud API documentation. Call this first before searching.

Inputs: max_pages. Returns: result

index_prisma_docsA

Index Prisma Cloud documentation. Call this first before searching.

Inputs: max_pages. Returns: result

search_all_docsC

Search across all Prisma Cloud documentation sites.

Inputs: query (required). Returns: result

search_prisma_api_docsC

Search Prisma Cloud API documentation

Inputs: query (required). Returns: result

search_prisma_docsC

Search Prisma Cloud documentation

Inputs: query (required). Returns: result

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    B3.4/5.0

    Scored across 6 tools

    Disambiguation3/5

    The tools have some overlap that could cause confusion, particularly between the three search tools (search_all_docs, search_prisma_api_docs, search_prisma_docs) and the two index tools (index_prisma_api_docs, index_prisma_docs). However, the descriptions clarify the scope of each, helping to distinguish them. The get_index_status tool is clearly distinct in purpose.

    Naming Consistency5/5

    All tool names follow a consistent verb_noun pattern with snake_case throughout, such as get_index_status, index_prisma_api_docs, and search_all_docs. This predictability makes it easy for agents to understand and use the tools without confusion.

    Tool Count5/5

    With 6 tools, the server is well-scoped for its purpose of indexing and searching Prisma Cloud documentation. Each tool serves a clear function, and the count is neither too sparse nor overwhelming, fitting typical expectations for a documentation-focused MCP server.

    Completeness4/5

    The tool set covers the core workflows of indexing and searching documentation, with separate tools for API and general docs, plus a status check. A minor gap is the lack of tools for managing or updating indexed content, but agents can work around this by re-indexing as needed.

    Maintenance

    ActivitySlowing
    ResponsivenessNo recent community issues

    Related MCP Connectors

    Related MCP Servers