How to install fff.nvim with lazy.nvim?

Add the plugin to your lazy.nvim config with a build function to download or build the binary, and set keymaps like 'ff' for find_files. The README provides a complete example with opts for debugging and layout.

FFF.nvim vs telescope.nvim: which is better?

FFF focuses on raw speed and memory-augmented search for AI agents and large repos, while telescope.nvim offers a wider plugin ecosystem and more general-purpose picking. Choose FFF if performance and AI integration are critical.

Does fff.nvim work with AI agents like Claude Code?

Yes, it includes an MCP integration with a bash script for installation, and charts show it reduces token usage by providing memory-enhanced file search, as noted in the README's AI section.

How to filter search results by git status in fff.nvim?

Use constraints like 'git:modified' in the search query, or enable git status text coloring in setup. The plugin highlights status in the sign column by default and allows negation with '!'.

Can fff.nvim handle very large repositories?

Yes, it's optimized for large codebases with features like time-budgeted grep and lazy syncing, demonstrated on the Linux kernel with 100k files, though it may require tuning max_threads and file size limits.

What is the memory-augmented search in fff.nvim?

It stores file open frecency and query combos in databases to boost scores for frequently accessed files, helping both AI agents and users find relevant results faster based on past behavior.

Open-Awesome

fff.nvim

MITRustv0.10.1

A fast, memory-enhanced fuzzy file search toolkit for AI agents and Neovim, with typo-resistant matching and git integration.

Visit Website GitHub

9.8k stars387 forks0 contributors

What is fff.nvim?

FFF is a fast fuzzy file search toolkit built for AI agents and Neovim. It provides typo-resistant file matching, live grep with multiple modes, and memory-enhanced result ranking to speed up code navigation and reduce AI token usage. The tool integrates deeply with git and offers cross-mode suggestions when searches return no results.

Target Audience

Neovim users seeking a performant, intelligent file picker, and developers building AI agents that need efficient, memory-augmented file search capabilities within codebases.

Value Proposition

Developers choose FFF for its exceptional speed on large repositories, its memory-driven ranking that learns from usage patterns, and its dual utility for both human and AI-driven workflows, all while maintaining a lightweight, configurable interface.

Overview

The fastest and the most accurate file search SDK for AI agents, Neovim, Rust, C, Python, Bun and NodeJS

Use Cases

Best For

Speeding up file navigation in large monorepos with 100k+ files
Reducing AI agent token consumption by providing smarter file search
Finding files with typos or partial names using fuzzy matching
Filtering search results by git status (e.g., only modified files)
Batch selecting files and sending them to Neovim's quickfix list
Switching between plain text, regex, and fuzzy grep modes on the fly

Not Ideal For

Teams using editors other than Neovim 0.10+
Projects requiring a standalone, editor-agnostic file search tool
Environments where downloading or building Rust binaries is restricted
Users who prefer minimal-configuration, drop-in plugins without Lua setup

Pros & Cons

Pros

Fast Fuzzy Matching

Uses Smith-Waterman scoring for typo-resistant search, demonstrated on large codebases like the Linux kernel with 100k files for high performance.

Memory-Augmented Ranking

Tracks file open history and query combos in databases to boost relevant results, reducing AI token usage by minimizing roundtrips and speeding up human searches.

Live Grep Modes

Supports plain text, regex, and fuzzy grep modes cyclable with Shift+Tab, with smart case and time budgets to prevent UI freezes during searches.

Git Integration

Highlights git status via sign column or optional text colors, and allows filtering searches with constraints like 'git:modified' for focused results.

Cons

Binary Dependency

Requires downloading or building a Rust binary during installation, adding complexity and potential failures in environments without internet or Rust toolchains.

Neovim-Only Lock-in

Tightly coupled to Neovim 0.10+, making it unusable for other editors or standalone CLI use, limiting its portability and broader adoption.

Configuration Overhead

Extensive setup with numerous Lua options, highlight groups, and keymaps, which can be overwhelming and time-consuming compared to simpler plugins.

Frequently Asked Questions

Related Projects

telescope.nvim

Find, Filter, Preview, Pick. All lua, all the time.

Stars19,653

Forks957

Last commit26 days ago

folke/snacks.nvim#picker

🍿 A collection of QoL plugins for Neovim

Stars7,900

Forks481

Last commit1 month ago

fzf-lua

Improved fzf.vim written in lua

Stars4,380

Forks259

Last commit4 hours ago

telescope-media-files.nvim

Telescope extension to preview media files using Ueberzug.

Stars523

Forks56

Last commit2 years ago

Community-curated · Updated weekly · 100% open source

Found a gem we're missing?

Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.

Submit a project Star on GitHub