📦 stack-opt

A CLI tool I wrote in Rust to help audit JavaScript/TypeScript dependencies. It scans your package.json and highlights packages that are:

• 🐘 unusually large
• 🕒 stale (haven’t been updated in a while)
• ⚠️ deprecated
• 👤 low-maintenance (only one maintainer, low activity)

The goal: make it easier to keep projects lean and avoid risky dependencies.

---

Why I Built This

I wanted to learn Rust by building something practical. Dependency sprawl is a pain point I’ve hit in real projects, so I thought: why not try making a tool that helps me spot problems early?

This project gave me hands-on experience with:

• Writing CLIs in Rust
• Working with async APIs (NPM registry)
• Balancing speed with usability (human-readable vs JSON output)

---

Features

• Fast, local analysis powered by Rust
• Checks both dependencies and devDependencies
• Fetches live metadata from the NPM registry
• Colorful, easy-to-read CLI output
• JSON output for CI pipelines or automation
• Zero tracking — everything runs locally

---

Installation

From crates.io:

cargo install stack-opt

Or build from source:

git clone https://github.com/tomc2154/stack-opt.git
cd stack-opt
cargo install --path .

---

Usage

Basic scan:

stack-opt --path ./your-project

CLI Options

Flag	Description
--path	Path to the project directory (default: .)
--no-dev	Skip devDependencies
--json	Output results in JSON format
--fail-on	Exit with code 1 on warnings or criticals

Examples:

stack-opt --no-dev
stack-opt --json > audit.json
stack-opt --fail-on crit

---

Example Output

📦 Fetching metadata for moment [dependencies]...
- [WARN] moment: 🐘 Large size (~4248.4 KB)

===========================
📊 Final Summary:
---------------------------
🔴 Critical: 1
🟠 Warnings: 9
🔵 Info:     4
✅ Clean:    6
📦 Total:    20
===========================

---

What I Learned

• Structuring a Rust CLI project with Cargo
• Designing user-friendly command-line interfaces
• Interacting with external APIs efficiently in Rust
• Balancing developer UX with raw performance

---

License

MIT © 2025 tommantonclery