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

@ohmyperf/mcp-server

v0.1.1

Published

MCP (Model Context Protocol) server exposing the ohmyperf engine as 'measure' and 'diff' tools for AI agents (Claude / OpenCode / Cursor / Cline / Continue).

Readme

@ohmyperf/mcp-server

MCP server for ohmyperf — exposes real-machine, real-browser web performance measurement to AI coding agents (Claude in OpenCode, Cursor, GitHub Copilot, etc.) as 12 tools and 7 prompts.

chrome-devtools-mcp lets an agent inspect a live browser. ohmyperf-mcp lets the agent measure, persist, diff, and enforce budgets — capabilities chrome-devtools-mcp structurally does not have.

Install

npm install -g @ohmyperf/mcp-server
# or use directly via npx
npx -y @ohmyperf/mcp-server

Requires Node ≥ 22. Playwright Chromium is downloaded on first measurement (~150 MB).

Wire into your AI agent

OpenCode (~/.config/opencode/opencode.json)

{
  "mcp": {
    "ohmyperf": {
      "command": "npx",
      "args": ["-y", "@ohmyperf/mcp-server"]
    }
  }
}

Claude Desktop (~/Library/Application Support/Claude/claude_desktop_config.json)

{
  "mcpServers": {
    "ohmyperf": {
      "command": "npx",
      "args": ["-y", "@ohmyperf/mcp-server"]
    }
  }
}

Cursor (.cursor/mcp.json)

{
  "mcpServers": {
    "ohmyperf": { "command": "npx", "args": ["-y", "@ohmyperf/mcp-server"] }
  }
}

What the agent gets

12 MCP tools

| Tool | Purpose | |---|---| | measure | Measure a URL with Playwright + CDP. Returns full Report JSON (CWV, audits, frame tree, resources). collectTrace=true adds long-task / render-blocking attribution. | | track_url | Longitudinal monitoring (ohmyperf-only). Measure + append to local NDJSON time series; returns trend verdict per metric. | | find_regression_cause | Causal attribution (ohmyperf-only). Compares two reports, returns ranked hypotheses (new render-blocking, grown assets, new long-tasks, new third-parties). | | diff | Mann-Whitney U significance test between two report.json files. | | diff_resources | Same as diff but accepts ohmyperf://reports/<file>.json URIs. | | enforce_budget | Contract-as-code (ohmyperf-only). Measure + evaluate against budget JSON; returns PASS/FAIL + exit code. | | analyze_report | Drill into one insight slice from a saved report (lcp-breakdown, long-tasks, third-parties, etc.) without dumping 50 KB JSON. | | list_runs | List saved reports in ~/.ohmyperf-mcp/reports/. | | list_styles | List the 4 brand styles (calibre / linear-app / stripe / vercel) with manifest metadata. | | generate_html_report | Render a saved report as a single-file HTML viewer. Writes to disk + returns path (avoids overflowing MCP response budgets). | | generate_deck | Render a saved report as a multi-slide HTML presentation. | | generate_markdown_summary | ~2 KB PR-comment-friendly Markdown of a saved report. |

7 MCP prompts

diagnose_report, compare_runs, suggest_fixes, audit_third_parties, check_budget, investigate_regression, monitor_trend — guided multi-tool flows for diagnosis, regression investigation, and longitudinal monitoring.

Storage

Reports are persisted at ~/.ohmyperf-mcp/reports/<measurementId>.json (each call to measure writes one). They are also exposed as MCP resources via ohmyperf://reports/<file>.json so ListResources can browse them.

Time-series points (from track_url) live at ~/.ohmyperf-mcp/timeseries/<sha256-of-url>.ndjson — append-only, one JSON object per line.

Why ohmyperf MCP and not chrome-devtools-mcp?

| Capability | chrome-devtools-mcp | ohmyperf-mcp | |---|---|---| | Live browser inspection | ✓ | — | | Persistent reports as MCP resources | — | ✓ | | Time-series tracking + trend detection | — | ✓ | | Causal regression attribution | — | ✓ | | Budget enforcement as exit-code primitive | — | ✓ | | Mann-Whitney U statistical diff | — | ✓ | | OOPIF (cross-origin iframe) coverage | partial | ✓ ~99% | | Self-contained HTML/deck artifact | — | ✓ | | Brand-aware reporting (calibre/linear/stripe/vercel) | — | ✓ |

The two MCP servers are complementary — run both for the strongest agent loop.

Verify locally

npx -y @ohmyperf/mcp-server <<< '{"jsonrpc":"2.0","id":1,"method":"initialize","params":{"protocolVersion":"2024-11-05","capabilities":{},"clientInfo":{"name":"manual","version":"0"}}}'

You should see a JSON-RPC initialize response on stdout.

License

Apache-2.0 — see LICENSE and NOTICE.

Links