npm package discovery and stats viewer.

Discover Tips

  • General search

    [free text search, go nuts!]

  • Package details

    pkg:[package-name]

  • User packages

    @[username]

Sponsor

Optimize Toolset

I’ve always been into building performant and accessible sites, but lately I’ve been taking it extremely seriously. So much so that I’ve been building a tool to help me optimize and monitor the sites that I build to make sure that I’m making an attempt to offer the best experience to those who visit them. If you’re into performant, accessible and SEO friendly sites, you might like it too! You can check it out at Optimize Toolset.

About

Hi, 👋, I’m Ryan Hefner  and I built this site for me, and you! The goal of this site was to provide an easy way for me to check the stats on my npm packages, both for prioritizing issues and updates, and to give me a little kick in the pants to keep up on stuff.

As I was building it, I realized that I was actually using the tool to build the tool, and figured I might as well put this out there and hopefully others will find it to be a fast and useful way to search and browse npm packages as I have.

If you’re interested in other things I’m working on, follow me on Twitter or check out the open source projects I’ve been publishing on GitHub.

I am also working on a Twitter bot for this site to tweet the most popular, newest, random packages from npm. Please follow that account now and it will start sending out packages soon–ish.

Open Software & Tools

This site wouldn’t be possible without the immense generosity and tireless efforts from the people who make contributions to the world and share their work via open source initiatives. Thank you 🙏

© 2026 – Pkg Stats / Ryan Hefner

archsetu

v0.1.0

Published

ArchSetu - Engineering Intelligence Platform

Readme

ArchSetu

Your coding agent is about to refactor a function with two hundred callers. It has no idea.

ArchSetu gives it eleven read-only tools to ask with, over MCP, running entirely on your machine.

claude mcp add archsetu --scope project -- npx -y @archsetuweb/cli@latest mcp .

That's the whole install. No account, no API key, no signup. Your code is never uploaded.

npm license


What the agent can ask

| Tool | Answers | | --- | --- | | repo_overview | What is this codebase? Health score, size, languages, entry points, risk counts. | | function_blast_radius | What breaks if I change this function? Callers, files spanned, which are entry points. | | file_blast_radius | What breaks if I change this file? Direct and transitive importers. | | explain_file | What does this one file do? Purpose, imports, exports, functions with line numbers. | | dead_code | What here is unreachable? | | complexity_hotspots | Where is this hard to change? | | security_findings | What looks unsafe? Secrets, unsafe eval, string-built SQL, and more. Secret values are masked. | | test_risk | What is risky and untested? | | dependency_hygiene | Why does this break in CI? Imports that are never declared in a manifest. | | ownership_risk | Who knows this code? Bus factor, contributors, churn. | | onboarding_guide | Where do I start reading? |

Every tool is read-only and declares itself as such, so a client can auto-approve them without granting write access. There is no tool that writes, moves or deletes.

The server reads only inside the directories you launch it with. Paths outside them are refused, symlinks resolved first, so a README in a dependency cannot talk your agent into reading ~/.ssh.

Works with Claude Code, Cursor, Claude Desktop, or any client that can launch a subprocess. Setup for each, plus every flag: packages/cli/README.md


Also a plain CLI

npx -y @archsetuweb/cli@latest analyze .      # full analysis
npx -y @archsetuweb/cli@latest health .       # just the health score
npx -y @archsetuweb/cli@latest dead-code .    # unreachable functions
npx -y @archsetuweb/cli@latest share          # publish a page about this project

Requires Node.js 20 or newer.


What it will not tell you

Read this before you trust a number. Each tool repeats the relevant limit inside its own results, so the agent sees it at the point of use rather than in documentation it never read.

  • It reads code, it does not run it. No tests run, no coverage measured. "Untested" means no test reaches the function through the static call graph, which is not the same as uncovered.
  • The call graph cannot see dynamic calls. Reflection, dynamic dispatch, DI, callbacks passed by name, and calls from outside the repo are invisible. Dead code is a list of candidates to verify; blast radius is a lower bound.
  • Calls resolve by function name. Where two functions share a name, a call is attributed to one of them. Distrust a surprising result for handle, run or init.
  • Security findings are regex heuristics. No AST, no taint analysis, no vulnerability database. Expect false positives in fixtures and example config; read the cited line before believing any of it.
  • Very large repositories are truncated. The file walk is capped, and results say so rather than presenting a partial analysis as a complete one.

This repository

A pnpm monorepo. The MCP server and CLI are the part most people want; the rest runs archsetu.com.

| Path | What it is | | --- | --- | | packages/cli | The archsetu CLI and MCP server, published as @archsetuweb/cli | | packages/core | The analysis engine: parsers, call graph, metrics | | apps/web | The hosted app at archsetu.com | | packages/github-action | Run an analysis in CI | | packages/gitlab-ci | The same, for GitLab | | packages/vscode-extension | Editor integration |

Building from source

pnpm install
pnpm --filter @archsetu/core build    # core must build before the CLI
pnpm --filter @archsetuweb/cli build

Architecture notes: ARCHITECTURE.md · Security posture: SECURITY.md, THREAT_MODEL.md


MIT licensed. Full documentation at archsetu.com/mcp.