Documentation: Getting Started · README · Configuration · IDE Clients · MCP API · ctx CLI · Memory Guide · Architecture · Multi-Repo · Kubernetes · VS Code Extension · Troubleshooting · Development

The open-source, self-improving code search that gets smarter every time you use it.

AI coding assistants are only as good as the context they retrieve. Most solutions chunk your code into large blocks and hope for the best—returning whole files when you need a single function, or missing the relevant code entirely. Context-Engine takes a different approach: ReFRAG-inspired micro-chunking returns precise 5-50 line spans, hybrid search combines semantic and lexical signals with cross-encoder reranking, and the system adapts to your codebase over time. No cloud dependency, no vendor lock-in—just a Docker Compose stack that works with any MCP-compatible tool.

git clone https://github.com/m1rl0k/Context-Engine.git && cd Context-Engine
docker compose up -d

That's it. The stack is running.

Option A: VS Code Extension (recommended)

Install Context Engine Uploader from the VS Code Marketplace. Open your project, click "Upload Workspace". Done.

The extension auto-syncs changes and configures your MCP clients.

Option B: CLI

# Index any project
HOST_INDEX_PATH=/path/to/your/project docker compose run --rm indexer

HTTP endpoints — for Claude Code, Windsurf, Qodo, and other RMCP-capable clients:

{
  "mcpServers": {
    "qdrant-indexer": { "url": "http://localhost:8003/mcp" },
    "memory": { "url": "http://localhost:8002/mcp" }
  }
}

stdio via npx (recommended) — unified bridge with workspace awareness:

{
  "mcpServers": {
    "context-engine": {
      "type": "stdio",
      "command": "npx",
      "args": [
        "@context-engine-bridge/context-engine-mcp-bridge",
        "mcp-serve",
        "--workspace", "/path/to/your/project",
        "--indexer-url", "http://localhost:8003/mcp",
        "--memory-url", "http://localhost:8002/mcp"
      ]
    }
  }
}

See docs/IDE_CLIENTS.md for Cursor, Windsurf, Cline, Codex, Augment, and more.

The Context Engine Uploader extension provides:

• One-click upload — Sync your workspace to Context-Engine
• Auto-sync — Watch for changes and re-index automatically
• Prompt+ button — Enhance prompts with code context before sending
• MCP auto-config — Writes Claude/Windsurf MCP configs for you

See docs/vscode-extension.md for full documentation.

Search (Indexer MCP):

• repo_search — Hybrid code search with filters
• context_search — Blend code + memory results
• context_answer — LLM-generated answers with citations
• search_tests_for, search_config_for, search_callers_for

Memory (Memory MCP):

• store — Save knowledge with metadata
• find — Retrieve stored memories

Indexing:

• qdrant_index_root — Index the workspace
• qdrant_status — Check collection health
• qdrant_prune — Remove stale entries

See docs/MCP_API.md for complete API reference.

flowchart LR
  subgraph Your Machine
    A[IDE / AI Tool]
    V[VS Code Extension]
  end
  subgraph Docker
    U[Upload Service]
    I[Indexer MCP]
    M[Memory MCP]
    Q[(Qdrant)]
    L[[LLM Decoder]]
    W[[Learning Worker]]
  end
  V -->|sync| U
  U --> I
  A -->|MCP| I
  A -->|MCP| M
  I --> Q
  M --> Q
  I -.-> L
  I -.-> W
  W -.-> Q

The VS Code extension syncs your workspace to the stack. Your IDE talks to the MCP servers, which query Qdrant for hybrid search. Optional features include a local LLM decoder (llama.cpp), cloud LLM integration (GLM, MiniMax M2), and adaptive learning that improves ranking over time.

Python, TypeScript/JavaScript, Go, Java, Rust, C#, PHP, Shell, Terraform, YAML, PowerShell

MIT