fff: File Search 100x Faster Than ripgrep

⚡ fff: the file finder that beats ripgrep

Do you use ripgrep or fzf to search your code? Meet fff (Fast File Finder), a Rust library that is 100x faster in long-running processes that search more than once.

🔍 What makes fff special?

• In-memory index: no regex, no disk index, instant search
• Frecency memory: files you actually open rank higher automatically
• Smart-case with fuzzy fallback: IsOffTheRecord finds snake_case variants; zero matches retries as fuzzy
• Git-aware: tags modified, untracked, and staged files
• Definition-first: code definition lines are classified on the Rust side, no prompt overhead

🤖 For AI agents (MCP server):

curl -L https://dmtrkovalenko.dev/install-fff-mcp.sh | bash

Works with Claude Code, Codex, Cursor, and any MCP-capable client. Fewer grep roundtrips, less wasted context.

🖊️ Also as a Neovim plugin — demo on the Linux kernel (100k files, 8GB).

💡 Explanation in a nutshell

#

ripgrep is great at searching files once. But when you use an editor or AI agent that searches constantly, fff keeps everything in memory and learns which files you use most, making each search progressively faster and more relevant.

GitHub - dmtrKovalenko/fff.nvim: The fastest and the most accurate file search toolkit for AI agents, Neovim, Rust, C, and NodeJS

The fastest and the most accurate file search toolkit for AI agents, Neovim, Rust, C, and NodeJS - dmtrKovalenko/fff.nvim

github.com ↗

More information at the link 👇

fff – The Future of Code Search

Blazing-fast file finder and content grep across massive repositories. Search Linux (38M loc), Chromium (42M loc), and more in milliseconds.

fff.dmtrkovalenko.dev ↗