Claude Code can now navigate your code like an IDE
Claude Code used to search a codebase with grep and text matching. With Language Server Protocol plugins it gets real code intelligence: references, signatures and types.
Written by
-
Benedikt Stemmildt Founder & Co-CEO
Published: 12 February 2026 Updated: 1 min read
Claude Code searches your codebase like a junior developer: grep, text matching, reading files one by one.
Meanwhile, every IDE on the planet uses the Language Server Protocol (LSP) to navigate code instantly. Now Claude Code can do the same.
What LSP changes
- Without LSP: naive text search across files. It misses references and guesses at function signatures.
- With LSP: proper code intelligence. It finds all references and reads exact parameter types and descriptions.
Setup takes one command
Type /plugin in Claude Code, pick your language, and you are done.
Python gets Pyright. TypeScript, Go and Rust are all supported out of the box. There is no MCP server to set up and no configuration file to write.
Why this matters
Professional developers don’t grep through codebases. They Ctrl+Click to jump to definitions, hover for type info, and use “Find All References” to understand the impact before changing anything.
Claude Code now works the same way:
- more accurate code modifications in large codebases
- better understanding of function signatures and types
- fewer hallucinated parameters and imports
- faster navigation, and fewer tokens wasted on reading files
Context engineering at the tool level
This is what context engineering looks like at the tool level. Better context in, better code out.
The Serena MCP server proved this concept with more than 17,000 GitHub stars. Now it is built in. That is the pattern: the community innovates, and platforms absorb.
The setup is shown in this video on LSP support in Claude Code (opens in a new tab).
Are you still letting your agent grep through files, or giving it real code intelligence?
Further reading
Written by
-
Benedikt Stemmildt · Founder & Co-CEO
More on Coding agent setup
- Agent teams win by moving complexity out of the agent Vercel, Manus and Pi got better results by removing tools from the agent. The orchestration belongs in deterministic factory steps around a small, focused agent.
- The best agent architecture is a simple while loop Claude Code and the pi agent show the same thing: a main loop, a few general-purpose tools and careful context management beat elaborate orchestration.
- Agents should write code to call MCP tools, not call them directly In Anthropic's example, direct MCP tool calls took 150,000 tokens and an agent writing code to call the same tools took 2,000. Code execution is the way to use MCP.
Scaling needs named owners
Coding agents scale across teams when named people own the platform, the team practice, the guardrails, and the measurement. Rollout & scale is where that ownership is built and handed to internal owners, in your stack.
Training on this topic: Agentic Engineering for Developers and QAWant to check one team first? The self-check: twelve yes or no questions, three minutes, instant result, no account. Take the self-check