parcadei

parcadei / fastedit

Public

AST-aware fast code editing via local 1.7B model

37
4
100% credibility
Found Apr 27, 2026 at 37 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

FastEdit is a command-line tool for efficiently editing code files using structure-aware analysis and a compact AI model to merge minimal change descriptions.

How It Works

1
🔍 Discover FastEdit

You hear about FastEdit from coding friends or online, a smart helper that makes changing code files super quick and saves tons of effort.

2
📦 Install with one command

Run a simple install command on your computer, and it sets everything up automatically without hassle.

3
📥 Grab the thinking brain

Download the small AI brain once, so your helper can understand and edit code fast on your machine.

4
🗺️ Peek at your code map

Ask to see the outline of functions and classes in your file, like a table of contents, to spot exactly what to change.

5
✏️ Describe your change

Type a short note on what to tweak, like 'add error handling here', and point to the spot with a name.

✅ Your code updates perfectly

Watch as your file gets edited precisely where you wanted, fast and accurate, ready for your next project step.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 37 to 37 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is fastedit?

Fastedit is a Python CLI for AST-aware code editing via a local 1.7B model, solving the token waste in AI tools that force models to repeat old code for diffs or SEARCH/REPLACE. Target functions/classes by name with tree-sitter, apply snippets instantly (74% deterministic, 0 tokens) or via fast merges (<1s), across 13 languages like Python, JS, Rust. Commands like `fastedit edit api.py --replace process --snippet 'try: ...'` handle batch edits, deletes, renames, moves.

Why is it gaining traction?

Cuts output tokens 44-54% vs Claude/Aider by ditching location repeats – model writes changes only. Local MLX/vLLM backends run on laptops/GPUs; MCP server plugs into Claude Code/Cursor for seamless agent use. Deterministic paths nail simple inserts/replaces free, model kicks in just for rewrites.

Who should use this?

Python/Django backend devs batch-refactoring APIs, JS/TS fullstackers renaming symbols across files, Rust/Go teams deleting unused funcs safely. Suits local AI coding on Apple Silicon without cloud latency, or MCP for agent-driven edits in Cursor/Claude.

Verdict

Download fastedit via `uv tool install fastedits[mlx,mcp]` – solid for fast, local AST-aware editing if you need token efficiency now. 37 stars and 1.0% credibility signal early stage, but install/docs are polished; test on your repo before production.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.