ContextOS

Overview

ContextOS is an MCP server that aims to be the operating system layer for AI context by absorbing, extending, and surpassing capabilities of seven leading open-source repositories. It provides a cognition layer with six cognitive primitives for reasoning between retrieval and generation, a retrieval router that handles data based on churn rate, and an index lifecycle manager for self-healing indexes. The platform includes orchestration features like semantic intent routing, request tracing, schema registry, and multi-workspace auth.

Try asking AI

After installing, here are 5 things you can ask your AI assistant:

• register_source

  Register a data source with specific churn profile and indexing strategy

  Try:Register a new data source for search queries with live churn classTry:Add a cold data source with vector indexing strategy for policy documents
• router

  Access the data source router for managing and retrieving information

  Try:Show me the registered data sources and their churn profilesTry:Retrieve information using the churn-aware routing system
• semantic_intent_router

  Classify incoming requests and dispatch to correct processing layer

  Try:Classify this request and route it appropriatelyTry:Determine if this task requires retrieval or synthesis
• request_tracing

  Track lineage and observability data for tool calls

  Try:Show me the tracing information for the last tool callTry:Identify which layers fired during processing
• active_forgetting

  Filter out retrieved context that degrades output quality

  Try:Apply active forgetting to this retrieved informationTry:Remove noise from the context before processing
• reasoning_depth_calibration

  Determine appropriate depth of thinking for a given problem

  Try:Calibrate reasoning depth for this complex analysisTry:Estimate how much thinking this problem deserves
• synthesis_detection

  Identify whether the task requires retrieval or reasoning

  Try:Detect if this is a synthesis or retrieval taskTry:Determine if more data would help or hurt this analysis
• unknown_unknown_sensing

  Detect when missing entire categories of information

  Try:Check for unknown unknowns in this analysisTry:Flag potential missing information categories
• productive_contradiction

  Handle conflicting data as useful signal rather than noise

  Try:Resolve contradiction between these data pointsTry:Analyze conflicting signals to find insights
• context_dependent_gravity

  Re-weight memory importance based on current question

  Try:Apply context-dependent gravity to this memoryTry:Re-weight these memories based on the current query
• rebuild_index

  Trigger rebuild of data indexes manually or automatically

  Try:Rebuild the index for the keyword data sourceTry:Check if indexes need rebuilding and update them
• embedding_drift_detection

  Detect and handle embedding model version mismatches

  Try:Check for embedding model drift in all indexesTry:Update indexes after embedding model change

Note: Tool names were inferred from feature descriptions and code examples in the README. While not explicitly listed as a tool section, the functionality descriptions and code examples clearly indicate these MCP tools.

Comparable tools

Installation

pip install contextos

{
  "mcpServers": {
    "contextos": {
      "command": "python",
      "args": ["-m", "contextos"]
    }
  }
}

FAQ

How is ContextOS different from a wrapper of other MCP servers?

ContextOS is not a wrapper but a platform that transforms other tools into modules running on top of it, providing additional orchestration and cognitive capabilities.

What are the six cognitive primitives?

Active Forgetting, Reasoning Depth Calibration, Synthesis Detection, Unknown Unknown Sensing, Productive Contradiction, and Context-Dependent Gravity.

Compare ContextOS with