Skip to main content
ClaudeWave

Spatial architecture canvas and context optimization OS for AI coding agents. AST-sliced task contexts (99.4% token reduction), persistent architecture graphs across sessions, and test-gated verification via MCP.

MCP ServersOfficial Registry3 stars0 forksJavaScriptMITUpdated today
ClaudeWave Trust Score
95/100
Verified
Passed
  • Open-source license (MIT)
  • Actively maintained (<30d)
  • Clear description
  • Topics declared
  • Documented (README)
Last scanned: 9/12/2026
Install in Claude Code / Claude Desktop
Method: NPX · github
Claude Code CLI
claude mcp add contextos -- npx -y github
claude_desktop_config.json (Claude Desktop)
{
  "mcpServers": {
    "contextos": {
      "command": "npx",
      "args": ["-y", "github"]
    }
  }
}
1. Run the command above in your terminal (Claude Code), or paste the JSON config into claude_desktop_config.json (Claude Desktop).
2. Replace any <placeholder> values with your API keys or paths.
3. Restart Claude. The MCP server and its tools appear automatically.
Use cases

MCP Servers overview

<div align="center">
  <img src="assets/logo.png" width="96" alt="ContextOS Logo" />
  <h1>ContextOS</h1>
  <p><strong>Spatial Architecture Canvas & Context Optimization Operating System for AI Coding Agents</strong></p>
  <p>
    AST-sliced task contexts (99.4% token reduction) · Cross-session architecture memory · Deterministic verification gates
  </p>
  <p>
    <a href="README_zh.md">中文文档</a> ·
    <a href="https://github.com/yubinbin32-ops/ContextOS/releases/latest"><strong>Download Native App (macOS 14+)</strong></a> ·
    <a href="#headless--cross-platform-cli">Headless CLI / Windows</a> ·
    <a href="#empirical-benchmarks">Empirical Benchmarks</a> ·
    <a href="https://glama.ai/mcp/servers/yubinbin32-ops/ContextOS">MCP Directory</a>
  </p>
  <p>
    <a href="https://github.com/yubinbin32-ops/ContextOS/releases"><img alt="GitHub release" src="https://img.shields.io/github/v/release/yubinbin32-ops/ContextOS?style=flat-square&color=111111" /></a>
    <a href="https://github.com/yubinbin32-ops/ContextOS/actions/workflows/release.yml"><img alt="Build status" src="https://img.shields.io/github/actions/workflow/status/yubinbin32-ops/ContextOS/release.yml?style=flat-square&label=build" /></a>
    <a href="LICENSE"><img alt="MIT license" src="https://img.shields.io/github/license/yubinbin32-ops/ContextOS?style=flat-square" /></a>
    <img alt="Node.js 22+" src="https://img.shields.io/badge/Node.js-22%2B-43853d?style=flat-square" />
    <img alt="MCP compatible" src="https://img.shields.io/badge/MCP-compatible-7c3aed?style=flat-square" />
    <img alt="Zero Runtime Deps" src="https://img.shields.io/badge/Runtime%20NPM%20Deps-0-brightgreen?style=flat-square" />
  </p>
</div>

<p align="center">
  <img src="assets/contextos-demo.gif" alt="ContextOS Spatial Canvas in action: filtering architecture blocks, tracking impact paths, and inspecting verification evidence" width="100%" />
</p>

<p align="center"><sub>Navigate spatial architecture → trace execution chains → deliver symbol locators to agents → verify deterministic evidence.</sub></p>

---

## 1. Abstract & Problem Statement

Modern AI coding agents (Claude Code, Cursor, Windsurf, Codex, Devin) encounter a structural bottleneck when scaled to medium-to-large software repositories: **Context Window Saturation and Architectural Entropy**.

1. **Flat Context Inefficiency**: Conventional agents indiscriminately ingest thousands of lines of raw source files to inspect individual methods. Over 85–95% of the attention budget is expended on boilerplate imports, formatting, and unrelated helper routines.
2. **Cross-Session Topological Drift**: Because context windows flush between prompts, agents lose the system's global architectural invariants. Decisions made in session A are violated in session B, generating cyclic regressions and architectural decay.
3. **Unverified Mutations**: Agents assert completion based on probabilistic self-assessment rather than deterministic evidence closure, bypassing integration contracts and test verification gates.

**ContextOS** solves this by establishing a dual-plane operating system:
- **The Spatial Canvas (Developer Interface)**: A native macOS SwiftUI workspace that projects software systems into interactive architectural blocks, directed dependency chains, and verification gates.
- **The Cognitive Plane (Agent Interface)**: A Model Context Protocol (MCP) server that delivers task-sliced AST symbol facades (`path::symbol`), enforces strict token budgets, and manages cryptographic test execution receipts.

---

## 2. Primary Distribution: Native macOS Spatial Workspace

> **ContextOS is primarily designed and distributed as a native macOS application** built with SwiftUI, Metal rendering, and embedded SQLite caching.

<p align="center">
  <img src="assets/canvas-overview.png" alt="Native ContextOS Canvas showing architecture blocks, orthogonal dependency paths, project groups, and the inspector" width="100%" />
</p>

### Native Workspace Capabilities:
- **Spatial Topology Engine**: Compact orthogonal dependency routing keeps complex architectures with 50+ modules readable and navigable.
- **Ghost-to-Solid Lifecycle**: Formulate new features as *Ghost Blueprints* before code exists; progressively anchor blocks to real AST symbols as implementations land.
- **Verification Gate Ledgers**: Live Checkpoint status indicators (e.g. `13/13 100% Passed`) backed by concrete test cases and compiler receipts.
- **Visual Impact Tracing**: Double-click any block or chain to illuminate one-hop dependencies, upstream callers, and downstream side-effects.
- **In-App Protocol Dispatch**: One-click registration and bundle synchronization for Codex, Cursor, Windsurf, and Claude Desktop.

<table>
  <tr>
    <td width="50%"><img src="assets/path-impact.png" alt="Selected impact path" /></td>
    <td width="50%"><img src="assets/checkpoint-detail.png" alt="Checkpoint evidence" /></td>
  </tr>
  <tr>
    <td align="center"><sub>Trace upstream & downstream impact before writing code.</sub></td>
    <td align="center"><sub>Inspect verifiable test receipts behind completion.</sub></td>
  </tr>
</table>

### 📥 Download Native App
Download the standalone application package directly from GitHub Releases:
- **[ContextOS for macOS (ContextOS-macos.zip)](https://github.com/yubinbin32-ops/ContextOS/releases/latest)**  
*(Requires macOS 14.0+. Distributed as a clean `.zip` application bundle — zero DMG translocation anomalies).*

---

## 3. Empirical Benchmarks

The following results were measured directly on a real-world repository (**54 Architectural Blocks, 8 Chains, 80 Directed Links, 19 Checkpoints**) using the reproducible benchmark suite (`npm run benchmark`):

| Evaluation Dimension | Baseline (Conventional File Ingestion) | ContextOS (AST Task Slice) | Empirical Delta |
| :--- | :---: | :---: | :---: |
| **Task Context Window** | 689,403 chars (~183,571 tokens) | 3,993 chars (~1,125 tokens) | **99.4% Token Reduction** |
| **4-Module Execution Chain** | 139,247 chars (~34,812 tokens) | 2,056 chars (~530 tokens) | **98.5% Token Reduction** |
| **Terminal Log Diagnostics** | 16,083 chars (~4,042 tokens) | 826 chars (~211 tokens) | **94.8% Compression** |
| **Context Retrieval Latency** | Sequential File Traversal | 29.68 ms (P50) | **Sub-30ms Instant Lookup** |
| **Topological Drift & Recall** | High Hallucination Risk | 100% Target Module Recall | **Zero Architectural Drift** |

To reproduce these metrics locally on your own machine:
```bash
npm run benchmark
```

---

## 4. Headless & Cross-Platform CLI

For headless CI/CD pipelines, remote servers, Windows, Linux, or users who do not require the visual desktop application, ContextOS runs headlessly via Node.js (≥22):

```bash
# 1. Initialize and automatically scan existing codebase topology
npx -y github:yubinbin32-ops/ContextOS init --scan

# 2. Inspect project architecture health, sync state, and verification gates
npx -y github:yubinbin32-ops/ContextOS status

# 3. Configure local MCP client integrations
npx -y github:yubinbin32-ops/ContextOS setup
```

---

## 5. Architectural Principles & Operational Closed Loop

```mermaid
flowchart LR
  Human["Developer\nNative Spatial Canvas"] <--> Plaintext[".contextos/graph.json\nGit-Tracked Truth"]
  Plaintext <--> Engine["Local SQLite Cache\nContext Engine"]
  Engine --> Slice["AST-Sliced Task Context\n(path::symbol locators)"]
  Slice --> Agent["AI Coding Agent\n(via MCP)"]
  Agent --> Evidence["Test Execution Receipt"]
  Evidence --> Gate{"Verification Gate"}
  Gate -->|Passed| Plaintext
  Gate -->|Failed| Alert["Drift Warning & Retest"]
```

### 1. AST Symbol Locators (`path::symbol`)
Instead of flooding the LLM context with full file dumps, ContextOS returns compact locators: target path, symbol signature, derived line boundaries, and interface contracts. The host editor opens only the target method.

### 2. Git-Native Plaintext Truth (`graph.json`)
The durable source of truth is a formatted, deterministic JSON file (`.contextos/graph.json`) versioned in Git alongside source code. A `git checkout` or `git revert` simultaneously restores code and architecture. An embedded SQLite engine provides zero-latency indexed queries with zero external runtime npm dependencies.

### 3. Receipt-Backed Checkpoints & Freshness Gating
Completion states cannot be asserted by AI declaration. They require execution receipts (`npm test`, compiler diagnostics) logged through `run_command` and bound via `checkpoint_record`. Any modification to bound source code automatically transitions dependent checkpoints to `retest_required`.

### 4. Terminal Log Sanitization
The command gateway intercepts terminal execution, strips ANSI sequences and progress bars, redacts local paths and secrets, and condenses repetitive logs into structured diagnostic summaries (94.8% token compression).

---

## 6. IDE & Agent Integration

ContextOS integrates natively via standard stdio Model Context Protocol (MCP).

### Configuration for Cursor, Windsurf, Claude Code, & Codex

Add to your MCP configuration file (e.g. `~/.cursor/mcp.json` or `claude_desktop_config.json`):

```json
{
  "mcpServers": {
    "contextos": {
      "command": "npx",
      "args": ["-y", "github:yubinbin32-ops/ContextOS", "serve"]
    }
  }
}
```

Or point directly to the bundled standalone engine:

```json
{
  "mcpServers": {
    "contextos": {
      "command": "node",
      "args": ["/absolute/path/to/contextos-mcp.mjs"],
      "env": {
        "CONTEXTOS_PROJECT_ROOT": "${workspaceFolder}"
      }
    }
  }
}
```

---

## 7. Local Development & Verification

ContextOS is built with zero external runtime npm dependencies:

```bash
# Clone the repository
git clone https://github.com/yubinbin32-ops/ContextOS.git && cd ContextOS

# Install build dependencies
npm ci

# Run the 55-test verification suite
npm test

# Run the empirical benchmark suite
npm run benchmark

# Build the MCP bundled server
npm run plugin:build

# Ve
agent-memoryai-codingai-coding-toolsantigravityarchitectureastclaude-codeclicodebase-intelligencecodexcodex-plugincontextcontext-engineeringcontext-managementdeveloper-toolsmacosmcpmodel-context-protocolsqliteswift

What people ask about ContextOS

What is yubinbin32-ops/ContextOS?

+

yubinbin32-ops/ContextOS is mcp servers for the Claude AI ecosystem. Spatial architecture canvas and context optimization OS for AI coding agents. AST-sliced task contexts (99.4% token reduction), persistent architecture graphs across sessions, and test-gated verification via MCP. It has 3 GitHub stars and its last recorded update is dated 2026-09-11.

How do I install ContextOS?

+

You can install ContextOS by cloning the repository (https://github.com/yubinbin32-ops/ContextOS) or following the README instructions on GitHub. ClaudeWave also provides quick install blocks on this page.

Is yubinbin32-ops/ContextOS safe to use?

+

Our security agent has analyzed yubinbin32-ops/ContextOS and assigned a Trust Score of 95/100 (tier: Verified). See the full breakdown of passed checks and flags on this page.

Who maintains yubinbin32-ops/ContextOS?

+

yubinbin32-ops/ContextOS is maintained by yubinbin32-ops. The last recorded GitHub activity is dated 2026-09-11, with 0 open issues.

Are there alternatives to ContextOS?

+

Yes. On ClaudeWave you can browse similar mcp servers at /categories/mcp, sorted by popularity or recent activity.

Deploy ContextOS to your cloud

Ship this repo to production in minutes. Each platform spins up its own environment with editable env vars.

Maintain this repo? Add a badge to your README

Drop the badge into your GitHub README to show it's tracked on ClaudeWave. Each badge links back to this page and reflects the live Trust Score.

Featured on ClaudeWave: yubinbin32-ops/ContextOS
[![Featured on ClaudeWave](https://claudewave.com/api/badge/yubinbin32-ops-contextos)](https://claudewave.com/repo/yubinbin32-ops-contextos)
<a href="https://claudewave.com/repo/yubinbin32-ops-contextos"><img src="https://claudewave.com/api/badge/yubinbin32-ops-contextos" alt="Featured on ClaudeWave: yubinbin32-ops/ContextOS" width="320" height="64" /></a>

More MCP Servers

ContextOS alternatives