惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

月光博客
月光博客
人人都是产品经理
人人都是产品经理
博客园 - 聂微东
WordPress大学
WordPress大学
S
SegmentFault 最新的问题
博客园 - Franky
V
V2EX
Y
Y Combinator Blog
Google DeepMind News
Google DeepMind News
J
Java Code Geeks
T
The Blog of Author Tim Ferriss
罗磊的独立博客
钛媒体:引领未来商业与生活新知
钛媒体:引领未来商业与生活新知
Jina AI
Jina AI
博客园 - 叶小钗
F
Fortinet All Blogs
让小产品的独立变现更简单 - ezindie.com
让小产品的独立变现更简单 - ezindie.com
A
About on SuperTechFans
M
MIT News - Artificial intelligence
云风的 BLOG
云风的 BLOG
Last Week in AI
Last Week in AI
D
Docker
博客园 - 【当耐特】
阮一峰的网络日志
阮一峰的网络日志

DEV Community

Authentication Security Deep Dive: From Brute Force to Salted Hashing (With Java Examples) Why AI Systems Don’t Fail — They Drift Spilling beans for how i learn for exam😁"Reinforcement Learning Cheat Sheet" I Replaced Chrome with Safari for AI Browser Automation. Here's What Broke (and What Finally Worked) How Python Borrows Other People's Work The $40 Architecture: Processing 1 Billion API Requests with 99.99% Uptime Vibe Coding: A Workflow Guide (From Zero to SaaS) Most webhook security guides protect the wrong side. The scary part is delivery. Headless CMS for TanStack Start: Build a Blog with Cosmic EU Age Verification App "Hacked in 2 Minutes" — What Actually Happened Comfy Cloud’s delete function does not actually remove files Running AI Models on GPU Cloud Servers: A Beginner Guide Event-driven media intelligence with AWS Step Functions and Bedrock I scored 500 AI prompts across 8 quality dimensions — here's what broke How to Call Google Gemini API from Next.js (Free Tier, No Backend Needed) The Portal Protocol: Reclaiming Human Connection in the Age of AI How to Fix Your Team's Scattered Knowledge Problem With a Self-Hosted Forum Intro to tc Cloud Functors: A Graph-First Mental Model for the Modern Cloud Designing Multi-Tenant Backends With Both Ownership and Team Access I Built a Neumorphic CSS Library with 77+ Components — Here's What I Learned PostgreSQL Performance Optimization: Why Connection Pooling Is Critical at Scale Cómo construí un SaaS multi-rubro para gestionar expensas en Argentina con FastAPI + Vue 3 🚀 I Built an Ethical Hacking Scanner Tool – Open Source Project I Replaced /usage and /context in Claude Code With a Single Statusline A Pythonic Way to Handle Emails (IMAP/SMTP) with Auto-Discovery and AI-Ready Design I Collected 8.9 Million Polymarket Price Points — Here's What I Found About How Markets Really Move EcoTrack AI — Carbon Footprint Tracker & Dashboard Everyone's Using AI. No One Agrees How. 5 self-hosted ebook managers worth trying in 2026 Building Your First AI Agent with LangChain: From Chatbot to Autonomous Assistant
I built a dependency health scanner in a day. Here's what...
varalaakshay · 2026-05-20 · via DEV Community

A few weeks back I inherited an old Node.js project and spent half a day grepping package.json trying to figure out which libraries were still alive. npm outdated told me which versions had updates. npm audit told me about CVEs. Neither told me what I actually needed to know: which of these packages have been quietly abandoned and what the community moved to.

So this past week I built one. It's called stack-rot, it's written in Python, and it's now on PyPI: pip install stack-rot.

This post is about what I shipped, what I cut, and the one decision that mattered more than the code.

What it does

Point it at a package.json and it tells you which dependencies are:

  • Abandoned — community has moved away, even if the registry doesn't say so (moment is the classic example)
  • Deprecated — officially flagged on npm or by maintainers (request, node-sass, tslint)
  • Healthy — actively maintained

For each problem dependency, it tells you why it's flagged, links to public evidence (a maintainer's announcement, an archived repo, a deprecation flag), and suggests alternatives.

Sample output on a real scan of Ghost:

🪦 ABANDONED (1 package):
  ❌ moment 2.24.0
     → In maintenance mode since 2020.
     → Migrate to: dayjs, date-fns, luxon

⚠️ DEPRECATED (2 packages):
  ❌ brute-knex 4.0.1 — Package no longer supported
  ❌ path-match 1.2.4 — Archived and no longer maintained

✅ HEALTHY (207 packages)
📊 Project health: 9.9/10

Enter fullscreen mode Exit fullscreen mode

Why a new tool

The honest version of the comparison table:

Tool What it focuses on
npm outdated Newer versions exist
npm audit Security vulnerabilities
Dependabot Automated version bumps
Snyk Security + license issues
Socket.dev Supply-chain risk
stack-rot Community migration intelligence

None of the existing tools answer the question developers actually ask when they open an inherited codebase: which of these packages should I stop using? That's the question stack-rot answers.

The two signals

Every scan combines:

  1. A hand-curated database (rot-db.json) of packages known to be dead, deprecated, or abandoned. Each entry has a status, a reason, suggested alternatives, and a public evidence URL. Started with 8 entries in v0.1. Every single one verified against the original source.

  2. Live npm registry data for every other package — uses the deprecation flag npm itself maintains.

The curated database wins when both fire. Human-verified evidence beats automated flags.

What I cut

The original plan called for shipping 50 curated entries on day one. I shipped 8.

Here's why. I had a list of about 50 "obviously dead" packages in my head — names every Node dev recognizes. I started writing entries for them. Around entry 10, I had a gut feeling about a couple and decided to web-search before writing more. Two of my "definitely dead" picks turned out to be wrong:

  • gulp — last in my head as a "task runner from the 2010s nobody uses anymore." Actually shipped Gulp 5.0.0 in March 2024 with breaking changes and active maintenance.
  • grunt — same vibe. Published a 1.6.2 release three weeks ago, under OpenJS Foundation governance, with around 3 million weekly downloads.

If I'd shipped those entries, the first user to run stack-rot against a project using gulp would have tweeted "your tool says gulp is dead but it's actively maintained" and credibility dies on day one.

So I cut from 10 to 8. The 8 I kept are all verified — moment, request, node-sass, tslint, bower, coffee-script, phantomjs, phantomjs-prebuilt. Each has a primary-source evidence URL. The other 42 can wait for community PRs that go through the same verification process.

The lesson: a small dataset that's 100% correct beats a large dataset that's 90% correct. Especially on day one.

What's on the roadmap

v0.1 is JavaScript only. The roadmap, in rough order:

  • v0.2 — Python (requirements.txt, pyproject.toml)
  • v0.3 — Automated abandonment signals (last-publish dates, repo activity)
  • v0.4 — Go (go.mod)
  • v0.5 — Rust (Cargo.toml)
  • v0.6 — Community sentiment data from public sources
  • v0.7 — JSON/HTML reports, CI mode, GitHub Action
  • v0.8 — Safe codemods for trivial migrations
  • v0.9 — Web dashboard and README badges
  • v1.0 — Stability and sustainability

The curated database grows by community contribution. Each PR has to include public evidence, working alternatives, and the right schema. Rules are in CONTRIBUTING.md.

Try it

pip install stack-rot
cd your-project
stack-rot

Enter fullscreen mode Exit fullscreen mode

Requires Python 3.10 or newer.

Source: https://github.com/varalaakshay-arch/stack-rot
PyPI: https://pypi.org/project/stack-rot/

If you've worked on dependency tooling or run a project that depends on something you suspect is abandoned, I'd love to hear from you — either on GitHub or in the comments here.