How to Use Sourcegraph Cody for AI Code Search
Leverage Sourcegraph Cody for codebase-wide AI search, code understanding, and intelligent navigation across large codebases.
Last updated:
What You'll Learn
This intermediate-level guide walks you through how to use sourcegraph cody for ai code search step by step. Estimated time: 10 min.
Step 1: Set up Cody with your codebase
Install Cody in your IDE and connect it to your Sourcegraph instance or use the cloud service with repository access.
Step 2: Use AI-powered code search
Search your entire codebase using natural language queries — ask how features work, find usage patterns, and locate relevant code.
Step 3: Get code explanations
Select any code and ask Cody to explain what it does, why it was written that way, and how it connects to other parts of the system.
Step 4: Generate code and tests
Use Cody to generate new code, unit tests, and documentation with awareness of your existing codebase patterns.
Step 5: Navigate large codebases
Use Cody's codebase understanding to navigate dependencies, find callers, and understand data flow across services.
Frequently Asked Questions
How does Cody's code search differ from GitHub search?▾
Cody understands code semantics and relationships, not just text matching. It can find code by what it does, not just what it contains.
Can Cody search across multiple repositories?▾
Yes. With Sourcegraph, Cody searches across all connected repositories simultaneously — essential for microservice architectures.
Is Cody worth it alongside Cursor?▾
Cody excels at code understanding and search across large codebases. Cursor excels at code generation and editing. They complement each other well.