PCP
Standardize Context, Not Intelligence. An open protocol and native MCP server for preserving, organizing, and serving structured project context (architecture, decision logs, domain rules, and roadmap) to AI coding assistants (Cursor, Claude Desktop, Antigravity) with built-in CLI verification
README
PCP (Project Context Protocol)
Standardize Context, Not Intelligence.
One Project Context. Every Contributor.
PCP (Project Context Protocol) is an open standard for preserving, organizing, and transferring project context so any contributor—human, AI, or automation—can continue a software project consistently, regardless of who or what takes over next.
Why PCP?
Software projects lose continuity every day.
Knowledge stays inside people's heads.
Important decisions disappear in chat histories.
Documentation becomes outdated.
AI assistants start from scratch because project context is fragmented.
New contributors spend valuable time rediscovering information that already exists.
As projects grow, context becomes scattered across documentation, source code, issue trackers, chats, and individual experience.
PCP exists to preserve project continuity.
🚀 Start Here
📖 New to PCP? Read our practical beginner guides:
⚡ Quickstart (CLI & MCP)
1. Initialize Context in Any Repository
npx @craftsolutions/pcp init --name "My Project"
2. Verify Context Integrity (Linter)
npx @craftsolutions/pcp check
3. Connect to AI Assistants via MCP (Cursor / Claude / Antigravity)
Add this to your IDE's MCP configuration:
{
"mcpServers": {
"pcp": {
"command": "npx",
"args": ["-y", "@craftsolutions/pcp", "mcp"]
}
}
}
🔍 Why PCP vs Other Solutions?
| Solution | Limitation | PCP Advantage |
|---|---|---|
| README.md | Surface-level landing page; lacks structured decision logs and rules. | 6 mandatory deep-context components. |
| Notion / Jira | External to code repository; AI cannot query it while coding. | Lives directly in the Git repository alongside code. |
| .cursorrules | Locked to Cursor; useless in Claude Desktop, VS Code, etc. | Universal open standard working across all AI tools. |
| MCP | MCP is the pipe (transport). | PCP is the standard context payload flowing through that pipe. |
Repository Guide
| Goal | Location |
|---|---|
| 🇹🇷 Türkçe Tanıtım Rehberi | docs/GUIDE_TR.md |
| 🇬🇧 Practical Guide | docs/PCP_EXPLAINED.md |
| Understand the vision | VISION.md |
| Understand the philosophy | MANIFESTO.md |
| Learn the principles | docs/CORE_PRINCIPLES.md |
| Learn the concepts | docs/CORE_CONCEPTS.md |
| Read the specification | docs/SPECIFICATION.md |
| Accepted RFCs | rfcs/RFC-0001-Core-Context-Model.md |
| Official CLI & MCP Server | cli/ |
| Copy the official template | reference/context-template/ |
| Explore a working example | examples/simple-project/ |
| See open work | docs/OPEN_WORK.md |
What is PCP?
PCP defines a shared Project Context standard.
Instead of teaching every AI system how your project works, PCP standardizes the information surrounding the project itself.
This enables:
- Humans
- AI assistants
- Autonomous agents
- Development tools
- Future contributors
to work from the same understanding of a project.
The Core Idea
Software projects should not depend on memory.
They should depend on shared context.
PCP creates a single, structured Project Context that survives:
- contributor changes
- AI model changes
- team growth
- long development pauses
- tooling evolution
One project.
One context.
Every contributor.
Core Principles
- Context over conversations
- Continuity over memory
- Concepts over files
- Open by design
- Human and AI collaboration
- Preserve knowledge, don't rediscover it
- Standardize context, not tools
How PCP Works
Contributor
│
▼
Project Context
│
▼
Next Contributor
│
▼
Project Continuity
Every contributor reads from the same Project Context.
Every contributor leaves the project with better context than they found.
What PCP Is
PCP is:
- An open standard
- A Project Context Protocol
- A continuity-first approach to software development
- Human and AI friendly
- Technology independent
- Implementation independent
What PCP Is Not
PCP is not:
- an AI framework
- an agent framework
- a prompt library
- a project management tool
- a documentation platform
- a version control system
- a knowledge base application
PCP standardizes context, not intelligence, tools, or workflows.
Reference Implementation
This repository contains the official reference implementation of PCP.
The reference implementation uses Markdown documents to demonstrate the concepts defined by the specification.
Alternative implementations may use different file formats, databases, or storage systems, provided they preserve the required concepts defined by the specification.
The specification defines concepts.
Reference implementations demonstrate practical implementations.
Repository Structure
docs/ Specifications and documentation
rfcs/ Design decisions and proposals
reference/ Official reference implementation
examples/ Example projects
assets/ Images and diagrams
scripts/ Development utilities
Project Status
⚠️ Active Design
PCP is currently in the specification phase.
The priority is building a stable conceptual foundation before developing tooling and integrations.
Roadmap
- Complete PCP Specification v0.1
- Publish Reference Context Template
- Expand example projects
- Build the Compliance Validator
- Develop the PCP CLI
- Publish VS Code integration
- Publish MCP integration
- Release PCP v1.0
Vision
We believe software projects should be able to continue regardless of who—or what—takes over.
Project continuity should not depend on memory.
It should depend on a shared, structured, and portable Project Context.
Philosophy
Don't preserve conversations.
Preserve context.
License
MIT License.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.