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

taskproof

v0.2.1

Published

CLI for running a matrix of real AI agents through task specs against your site

Vulnerabilities

Powered byPowered by Snyk
  • 0High
  • 0Medium
  • 0Low

Links

Git

Maintainers

cbetzcbetz

Keywords

ai-agentsagent-testingplaywrightllmcomputer-usebrowser-usecievalsweb-testing

Readme

taskproof

The taskproof command line tool: run a matrix of real AI agents (Claude computer-use, browser-use) through defined tasks on your site, score whether they actually complete each task, pinpoint the exact failing step and its cost, and diff agent-usability across releases.

Pre-release (0.x). The task-spec format and run-artifact schema may change between releases.

Install

npm install -g taskproof      # or run ad hoc: npx taskproof <command>
taskproof install-browsers    # fetch the Chromium the Claude adapter drives

Commands

  • taskproof init --url <site> — detect your framework and scaffold starter task specs.
  • taskproof validate <files...> — validate task-spec YAML against the schema (free, no API calls).
  • taskproof run <files...> --models <list> — drive each agent through each task, grade with pass@k, print a pass/fail matrix with cost per cell, and write run artifacts. Flags: --models, --max-cost (soft per-run cap), -k/--runs, --timeout, --headed, --no-judge, --out.
  • taskproof report --dir <runs> — render a self-contained HTML report (the matrix plus per-step traces with screenshots). --max-image-width downscales screenshots for a smaller, hostable report.
  • taskproof baseline save / taskproof diff — snapshot a known-good run, then catch regressions in CI. diff uses unix-diff exit codes (0 = clean, 1 = regression, 2 = error).
  • taskproof install-browsers — install the Chromium the Claude computer-use adapter needs.

Quickstart

taskproof init --url https://your-site.com
# edit taskproof/tasks/*.yaml — the goal and the success assertions
export ANTHROPIC_API_KEY=sk-ant-...
taskproof run taskproof/tasks/*.yaml --models claude-opus-4-8
taskproof report --dir taskproof-runs && open taskproof-runs/report.html

Writing task specs

A task spec is versioned YAML: a natural-language goal, an entryUrl, and deterministic url/dom/network assertions that decide whether a run passed. The format, every field, and the assertion types are documented in @taskproof/spec. taskproof init scaffolds editable starters; the defaults are SaaS-shaped (homepage CTA, pricing, trial), so reword the goal and assertion to fit your site.

More

The full walkthrough (CI/GitHub Action setup, the browser-use lane, and the fidelity-gap caveats) is in the project README and docs/TESTING.md. Apache-2.0.