Contextos

v1.0.0Knowledge & Memorystable

Unified MCP context intelligence platform — pip-installable CLI that absorbed 6 foundational repos. Context engineering for AI agents.

ai-agentsclaudeclicontext-engineeringcontext-management
Share:
22
Stars
0
Downloads
0
Weekly
0/5

What is Contextos?

Contextos is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to unified mcp context intelligence platform — pip-installable cli that absorbed 6 foundational repos. context engineering for ai agents.

Unified MCP context intelligence platform — pip-installable CLI that absorbed 6 foundational repos. Context engineering for AI agents.

This server falls under the Knowledge & Memory category on MCPgee, the world's largest MCP server directory with 33,000+ servers.

Features

  • Unified MCP context intelligence platform — pip-installable

Use Cases

Unified context intelligence
Context management and engineering
AI agent context orchestration
LicenseNOASSERTION
Languagepython
Versionv1.0.0
UpdatedApr 12, 2026
Statushealthy
Maintenanceactive

Works with

ClaudeOpenAIwindowsmacoslinux

Installation

Manual Installation

npx contextos

Configuration

Configuration Details

Config File

claude_desktop_config.json

Performance

Response Metrics

Response Time< 200ms
ThroughputMedium

Resource Usage

Memory UsageLow
CPU UsageLow

How to Set Up and Use Contextos

ContextOS is a unified context intelligence platform for AI agents that combines 67 MCP tools spanning memory storage and retrieval, cognition, planning, tool execution, and document intelligence into a single pip-installable package. It absorbed six foundational repositories into one CLI-driven system, making it possible for agents to store observations, perform hybrid retrieval across docs and web sources, detect contradictions in their knowledge, create and spar over plans, and route tasks across tool providers — all with a churn-aware routing layer that adapts to stale or unavailable data sources.

Prerequisites

  • Python 3.9 or later with pip
  • An MCP-compatible client such as Claude Desktop or a custom agent framework
  • Optional: Composio account if you want the composio tool integration
  • Optional: OpenAI or Anthropic API key if you want LLM-backed cognition features
1

Install ContextOS from PyPI

Install the package using pip. This makes the ctx CLI available and installs all 67 MCP tools.

pip install contextos
2

Initialise a ContextOS workspace in Python

Create a launch script that configures the workspace. Adjust memory_tier, retrieval_mode, and feature flags to match your agent's needs.

from contextos import ContextOS

ctx = ContextOS(
    workspace="my-agent",
    memory_tier="warm",
    retrieval_mode="hybrid",
    tools=["composio", "mcp"],
    sparring_hook=True,
    cognition_enabled=True,
    churn_aware_routing=True,
)
ctx.serve(port=8080)
3

Verify with the health check

After starting the server, confirm all subsystems report healthy before connecting a client.

ctx health
4

Add the server to your MCP client config

Point your MCP client at the running ContextOS server. Use the HTTP transport since it serves on a port.

{
  "mcpServers": {
    "contextos": {
      "command": "python",
      "args": ["launch_contextos.py"],
      "env": {
        "CONTEXTOS_WORKSPACE": "my-agent"
      }
    }
  }
}
5

Explore tools via the ctx CLI

Use the CLI to test memory storage, document search, and planning before connecting your AI agent.

ctx memory store "FastAPI requires async def with response_model"
ctx docs search "how to handle authentication"
ctx plan create "Build a REST API with JWT auth"
ctx cognition think --query "Which approach should I use for rate limiting?" --domain "backend"

Contextos Examples

Client configuration

Claude Desktop config entry launching ContextOS as an MCP server via a Python launch script.

{
  "mcpServers": {
    "contextos": {
      "command": "python",
      "args": ["/Users/yourname/agents/launch_contextos.py"],
      "env": {
        "CONTEXTOS_WORKSPACE": "my-agent"
      }
    }
  }
}

Prompts to try

Example agent prompts that exercise ContextOS memory, retrieval, and planning capabilities.

- "Store this architectural decision: we use PostgreSQL for all relational data"
- "Search my stored docs for anything related to database migrations"
- "Create a plan for adding OAuth2 support to the API and spar with yourself on the approach"
- "Check if there are any contradictions in what I've told you about the auth system"
- "What do you know about rate limiting from my stored memory?"

Troubleshooting Contextos

ImportError when running the launch script after pip install

Ensure you are running the same Python interpreter that installed the package. If you have multiple Python versions, use 'python3 -m pip install contextos' and 'python3 launch_contextos.py' to keep them consistent, or use a virtual environment.

ctx health shows a subsystem as unhealthy

Run 'ctx health' with verbose output to see which subsystem is failing. Retrieval failures are often caused by missing API keys for web or doc sources. Set the relevant API key environment variables before starting the server.

Memory tools store data but retrieve returns empty results

ContextOS uses a warm memory tier by default which may require indexing time. Call 'ctx indexer rebuild' to force a reindex of the stored memory, then retry the retrieval query.

Frequently Asked Questions about Contextos

What is Contextos?

Contextos is a Model Context Protocol (MCP) server that unified mcp context intelligence platform — pip-installable cli that absorbed 6 foundational repos. context engineering for ai agents. It connects AI assistants to external tools and data sources through a standardized interface.

How do I install Contextos?

Follow the installation instructions on the Contextos GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.

Which AI clients work with Contextos?

Contextos works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.

Is Contextos free to use?

Yes, Contextos is open source and available under the NOASSERTION license. You can use it freely in both personal and commercial projects.

Browse More Knowledge & Memory MCP Servers

Explore all knowledge & memory servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.

Quick Config Preview

{ "mcpServers": { "contextos": { "command": "npx", "args": ["-y", "contextos"] } } }

Add this to your claude_desktop_config.json or .cursor/mcp.json

Read the full setup guide →

Ready to use Contextos?

Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.

33,000+ ServersFree & Open SourceStep-by-Step Guides