Skip to main content

​
What Are Context Services?

Context Services provide high-quality semantic search to AI agents and applications. Whether you’re building custom agents or integrating with existing AI tools, Context Services give you access to Augment’s world-class context engine.

Context Engine MCP

Plug Context Engine into any agent via the Model Context Protocol

Context Engine SDK

Build context-aware custom agents or apps using our TypeScript or Python SDK

Context Connectors

Index any source and make it searchable via CLI, MCP, or HTTP

​
Context Engine MCP

Plug Context Engine into any agent, such as Claude Code, Codex, Gemini CLI, and more via the Model Context Protocol (MCP). Perfect for:
  • Using Augment’s context in your favorite AI tools
  • Enriching any coding agent or AI with state-of-the-art codebase semantic search
Get started with MCP →

​
Context Engine SDK

Build context-aware custom agents or apps using our TypeScript or Python SDK. The SDK provides FileSystem Context to index local directories or Direct Context to index files from any source. Perfect for:
  • Building custom AI agents with codebase understanding
  • Creating applications that need to understand codebase, documentation, specifications, runbooks, and other text sources
  • Achieving state-of-the-art context quality in AI-powered code review, security scanning, failure analysis, migrations, and more …
Get started with the SDK →

​
Context Connectors

An open-source library built on the Context Engine SDK that makes diverse sources searchable across agents and apps. Index code, documentation, runbooks, schemas, and configs from GitHub, GitLab, BitBucket, websites, or local filesystem. Store indexes locally or in S3. Search via CLI, interactive agent, or MCP server. Perfect for:
  • Indexing repositories without cloning them locally
  • Building applications with persistent indexes stored in S3
  • Auto-indexing on every push via webhooks
  • Indexing documentation websites
Get started with Context Connectors →