Perforce's P4mcp Server is an MCP server designed for integrating AI models and assistants with the Perforce P4 version control system. Built with Python, it uses FastMCP and direct P4 Python bindings to connect to P4 servers. It supports both STDIO and HTTP transport protocols, making it compatible with various AI clients like VS Code, Copilot, and IntelliJ IDEA. This server extends P4 capabilities, allowing developers and teams to incorporate AI into their daily development and code review workflows.

What does P4mcp Server connect to?

P4mcp Server connects directly to the Perforce P4 version control system. It functions as a Model Context Protocol (MCP) server, bridging AI applications with P4's data and operations. The server is built using Python, incorporating FastMCP and specific P4 Python bindings. For those building from source or using the uvx installation, Python 3.11 or newer is a requirement. This connection enables AI clients to interact with Perforce entities and execute commands.

Core features and tools

The server offers structured read/write tools for Perforce entities such as changelists, files, shelves, workspaces, jobs, reviews, streams, and server metadata. The query_server tool is always available. When search transforms are active, it exposes regex_search_tools, regex_call_tool, semantic_search_tools, and semantic_call_tool. It also supports P4 Code Review functions, including review discovery, voting, state transitions, commenting, and participant management, enhancing collaboration within P4.

Who is it for?

P4mcp Server is intended for developers, teams, and administrators who use the Perforce P4 version control system and want to integrate AI models and assistants into their development workflows. It helps these users bring AI capabilities into tasks like gaining operational clarity, assisting with code reviews, and automating development tasks. This server is particularly useful for environments already invested in the Perforce ecosystem. Skip this if your organization does not use Perforce P4 for version control.

How to get started

The server authenticates using valid P4 login credentials, which can be ticket-based or password-based. Setup involves establishing a connection to a P4 Server and configuring P4 environment variables such as P4PORT, P4USER, and P4CLIENT, or utilizing a P4CONFIG file. The server supports STDIO and HTTP transport protocols for client communication.

Use cases

• Gaining Operational Clarity: Users can instantly access code and assets, gather project and file-level activity and status, and retrieve complete version history through AI interfaces.
• Code Review Workflows: The server supports review discovery, voting, state transitions, commenting, and participant management, and can recommend reviewers.
• AI-assisted Development & Automation: It enables AI-assisted workflows for tasks like smart conflict resolution, creating automated checkpoints via intelligent shelving, and performing write operations such as creating changelists or checking out files.

P4mcp Server extends Perforce P4 functionality by connecting it with AI models and assistants. It provides tools for various P4 entities and supports critical development workflows, including code review and AI-assisted automation. The server's authentication relies on standard P4 credentials, with administrative controls for access. It is a valuable addition for Perforce users seeking to enhance their development processes with AI.