首页AI 工具Sourcegraph Cody
Sourcegraph Cody logo
Sourcegraph Cody
An AI coding assistant with deep codebase context, built on Sourcegraph's code search engine.
0
收藏数
Free (Pro); Enterprise custom
起步价
Sourcegraph
开发者
Sourcegraph Cody
产品截图

Sourcegraph Cody 是什么?

内容最后核实于 May 4, 2026 · 由 AI Tools Set 研究团队撰写

Cody is Sourcegraph's AI coding assistant, built on top of Sourcegraph's existing code search and intelligence engine — the thing that differentiates it from most code assistants is how well it understands an entire large codebase, not just the file currently open in your editor. It can answer questions about how a system works across multiple repositories, generate code with awareness of existing patterns in your codebase, and explain unfamiliar code.

It's built for teams working in large, complex codebases — especially enterprises with multiple repositories and legacy systems — where a coding assistant that only sees the open file misses too much context to be genuinely useful. Cody integrates with popular IDEs and supports multiple underlying LLMs rather than being locked to one model provider.

Cody Free is available for individual developers with generous usage limits. Enterprise pricing is custom, aimed at organizations that need Sourcegraph's broader code intelligence platform alongside Cody itself.

Sourcegraph Cody 的核心功能

  • Whole-codebase context: Understands code across multiple repositories, not just the currently open file.
  • Code search integration: Built on Sourcegraph's code search engine for finding and understanding existing patterns.
  • Multi-LLM support: Not locked to a single model provider — can work with multiple underlying LLMs.
  • Code explanation: Explains unfamiliar or legacy code in plain language.

Sourcegraph Cody 的使用场景

  • Understand how a feature works across a large, multi-repository codebase
  • Get AI-assisted code suggestions that respect existing codebase patterns
  • Explain unfamiliar legacy code to a new team member
  • Search and navigate a large codebase alongside AI-assisted coding

优缺点

优点
Whole-codebase context is a real differentiator for large, complex projects
Multi-LLM support avoids lock-in to a single model provider
Built on a mature code search platform, not a bolted-on chat feature
缺点
Most valuable for large, multi-repository codebases — less differentiated for small solo projects
Enterprise pricing isn't publicly listed, requiring a sales conversation
Setup for full codebase indexing takes more effort than a lightweight single-file assistant
在您的网站上推广 Sourcegraph Cody
分享
用户评价
Sign in to review
4.8
★★★★★
2,341 reviews
5
1,826
4
328
3
117
2
47
1
23
Sourcegraph Cody Review 2026: AI Coding Assistant with Codebase Context