kidowkidow/skills/find-npm-package/SKILL.md

find-npm-package

Skill2 starsChanged 6 months agoMITView on GitHub

Search npm packages by functionality using the npm registry API, iteratively refining the query until the user finds the right one. Covers JavaScript and TypeScript packages only — does not cover pip, cargo, gems, or other package registries. Use when user wants to find an npm package, node module, or JavaScript/TypeScript library for a specific purpose, or says things like "find me an npm package for X", "is there a node module that does Y", or "recommend a JS library for Z".

In this repository17 other files it gives its agents.
- create-nextjs-specificationSkill
- create-product-demoSkill
- find-github-repoSkill
- image-styleSkill
- nextjs-route-scaffoldSkill
- perfect-promptSkill
- place-meSkill
- review-meSkill
- setup-aeo-geoSkill
- setup-harnessSkill
- songwritingSkill
- step-by-stepSkill
- stock-analysisSkill
- study-meSkill
- teach-meSkill
- three-linesSkill
- visualstormingSkill

The file

Create a free account to mount

Files

SKILL.md

npm Package Finder

Quick Start

Extract 2–4 keywords from user's description, then search:

# With jq (preferred)
curl -s "https://registry.npmjs.org/-/v1/search?text=<keywords>&size=10" \
  | jq -r '.objects[] | "- **\(.package.name)** — \(.package.description // "(no description)") (score: \(.score.final * 100 | round)/100 · updated \(.package.date[:7]))\n  https://www.npmjs.com/package/\(.package.name)"'

# Fallback: python3
curl -s "https://registry.npmjs.org/-/v1/search?text=<keywords>&size=10" \
  | python3 -c "
import json, sys
data = json.load(sys.stdin)
for obj in data['objects']:
  p = obj['package']
  score = round(obj['score']['final'] * 100)
  updated = p['date'][:7]
  print(f\"- **{p['name']}** — {p.get('description', '(no description)')} (score: {score}/100 · updated {updated})\")
  print(f\"  https://www.npmjs.com/package/{p['name']}\")
"

Then fetch weekly downloads for the top 5 results:

# Replace <pkg> with each package name
curl -s "https://api.npmjs.org/downloads/point/last-week/<pkg>" \
  | jq -r '.downloads'

Append · Nk/week to each of the top 5 results.

Workflow

Extract keywords — pick nouns/verbs most specific to the desired functionality
Search via npm registry API
Fetch weekly downloads for top 5 packages
Present results as list (see Output Format)
Ask with choices:
"Did any of these match? A) Wrong runtime (need browser-only / Node-only / universal) B) Too generic — need something more specific C) Wrong category (utility / framework / CLI) D) Let me describe what I need E) Show more results"
Refine based on answer → repeat from step 2
No limit on rounds — keep going until user confirms they found it

Refinement Strategy

Choice	Action
A	Append "browser" / "node" / "universal" to query
B	Add more specific terms to query
C	Append "library" / "cli" / "framework" to query
D	Re-extract keywords from user's new description
E	Re-run with `size=20`, show next 10 results

API Options

# Basic search (sorted by npm score by default)
curl -s "https://registry.npmjs.org/-/v1/search?text=<query>&size=10"

# More results
curl -s "https://registry.npmjs.org/-/v1/search?text=<query>&size=20"

# Offset for pagination
curl -s "https://registry.npmjs.org/-/v1/search?text=<query>&size=10&from=10"

# Weekly downloads
curl -s "https://api.npmjs.org/downloads/point/last-week/<pkg>"

Output Format

Present results as:

package-name — description (score: N/100 · Nk/week · updated YYYY-MM)
https://www.npmjs.com/package/package-name

Sort by npm score descending (API default). Weekly downloads on top 5 only. If description is empty, note "(no description)". Format date as YYYY-MM.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

It worked
Partly
It didn't

What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free