Ganglia

What is Ganglia?

Ganglia indexes your project into a call graph and exposes 73 semantic tools for code analysis and manipulation. It integrates with coding assistants like Claude Code, Cursor, Windsurf, Continue, Cline, and Goose, and supports TypeScript, JavaScript, Rust, Python, Go, and PHP.

How to use Ganglia?

Ganglia is used as an MCP server added to supported coding assistants. Users configure it within their assistant’s tool settings to access the 73 available semantic tools.

Key features of Ganglia

• Indexes projects into a complete call graph
• Provides 73 semantic tools including code_get, code_callers, code_impact, code_propagate, and code_apply_batch with one-click revert
• Offers code_safe_move, deliberation chain, smart LLM-backed digests, and cross-session memory
• Reduces token usage by 80–95% compared to raw file-reading
• Works with Claude Code, Cursor, Windsurf, Continue, Cline, and Goose
• Supports TypeScript, JavaScript, Rust, Python, Go, and PHP
• Free during public beta

Use cases of Ganglia

• Navigating and understanding complex codebases through call graph queries
• Analyzing the impact of code changes before applying them
• Safely moving or refactoring code with one-click revert
• Generating smart, concise digests of code sections for faster comprehension
• Running batch semantic operations across a project

FAQ from Ganglia

What languages does Ganglia support?

Ganglia supports TypeScript, JavaScript, Rust, Python, Go, and PHP.

How does Ganglia reduce token usage?

Ganglia achieves 80–95% token savings compared to reading raw files by indexing the project into a call graph and providing semantic tools that deliver only relevant information.

Is Ganglia free?

Yes, Ganglia is free during its public beta period.

Which coding assistants can I use Ganglia with?

Ganglia works with Claude Code, Cursor, Windsurf, Continue, Cline, and Goose.

What tools does Ganglia provide?

Ganglia offers 73 semantic tools including code_get, code_callers, code_impact, code_propagate, code_apply_batch (with one-click revert), code_safe_move, deliberation chain, smart LLM-backed digests, and cross-session memory.