Phyll

io.github.carlosphyllPublished by GitHub account carlosphyllv0.4.6Updated Oct 25, 2026

UX review for apps built with AI, run by your own agent. The fixes keep your design.

VerifiedSTDIODesktop onlyBrowser AutomationDeveloper ToolsData & Analytics

Overview

AI-generated overview

Phyll gives an AI agent a browser and review method to run UX reviews of AI-built apps and write a report of what blocks first-time users.

What you can do
Start a review: scan the project, create the report folder and return the review method, with guide returning parts of it on demand and finish_review sending the report to the engine and writing report.md to disk. Capture pages: save screenshots and page probes of each route at laptop and phone size. Drive the browser: open pages, take snapshots, click, fill, select, check, press, go back, screenshot, probe and resize, staying on your app. Scan source for AI tells, with JSON or README badge output, and check account or upgrade plan.
When to use it
Useful when you have built an app with AI assistance and want your own agent to walk its core jobs like a first-time user and report friction such as sign-up walls, dead buttons or faint text, while keeping your design.
Requirements
Runs locally over stdio as an npm package; Node 20 or newer. Setup registers the connector in Codex, Claude Code, Cursor, Windsurf or Gemini CLI and installs a Chromium build. A Phyll account is needed for reviews; signup takes an email and saves the key in ~/.phyll, and login accepts an existing key. Network access to Phyll's engine.
Before you install
Reviews send data off your machine: the app address, project name and a scan summary (tells found, counts, route and form paths) at start, and the finished report at the end, which the engine keeps. Source code, screenshots and the agent conversation stay local. The account key is stored in ~/.phyll. Free signup includes 5 full reviews; Phyll Pro is a paid monthly subscription, and billing commands change the card or cancel.

Installation

In SourceWeft

  1. Open Phyll in the dashboard and add it to a workspace.
  2. Enable the server for the chats that should use its tools.

Desktop only via STDIO. STDIO servers start a local process, so they need the SourceWeft desktop host.

Other MCP clients

Follow the launch instructions in the repository.

README

phyll

UX review for apps built with AI, inside the agent you already use.

Phyll plugs into Codex, Claude Code or any agent that speaks MCP. Your agent opens your app the way a first-time user would, walks the core jobs in a real browser, and writes a report of what gets in the way: forms that ask for more than the job needs, sign-up walls, dead buttons, text too faint to read. The fixes keep your design.

The AI work runs in your agent, on your own plan. This package adds the browser, the page probe, the source scanner and the link to Phyll's engine, which sends the review method, checks the report and keeps it with a link.

Start

bash
npx phyll signup [email protected]npx phyll setup codex        # or claude, cursor, windsurf, gemini

Then ask your agent: "review my app at http://localhost:3000".

setup registers the connector in Codex's config.toml, with claude mcp add, or in the mcpServers of Cursor's ~/.cursor/mcp.json, Windsurf's ~/.codeium/windsurf/mcp_config.json or Gemini CLI's ~/.gemini/settings.json, keeping everything else in the file. It also installs the Chromium build the connector uses. For any other agent that takes MCP servers, npx phyll setup other prints the entry to paste. You need Node 20 or newer.

Your reports, keys and plan are also on the site: npx phyll account opens your account there, already signed in. If you lose the key, sign in at agentphyll.com/login with your email and create a new one.

Commands

CommandWhat it does
signup <email>Create a free account, with 5 full reviews. The key is saved in ~/.phyll
login <key>Use an account you already have on this computer
setup <agent>Connect Phyll to codex, claude, cursor, windsurf or gemini and install the browser; setup other prints the settings for any other agent
statusYour plan and the reviews left
accountOpen your account on agentphyll.com, already signed in: reports, usage, keys and plan
proSubscribe to Phyll Pro, R$ 9 a month, with unlimited reviews
billingChange the card or cancel Phyll Pro
scan [folder]Scan the source for AI tells, with no account and no AI. --format json for scripts, --format badge for a README badge
mcpRun the connector for your agent; setup registers it for you
logoutForget the key on this computer

The tools your agent gets

start_review scans the project, creates the report folder and returns the method. capture saves a screenshot and the page probe of each route at laptop and phone size. open, snapshot, click, fill, select, check, press, back, screenshot, probe and resize drive the browser, which stays on your app. guide returns a part of the method, finish_review sends the report to the engine and writes report.md, and scan, account and upgrade do what their names say.

What leaves your computer

Your source code, your screenshots and your agent's conversation stay on your computer. When a review starts, the connector sends the app's address, the project name and a summary of the scan: which tells it found, how many times, and the paths of routes and forms. When it ends, it sends the report your agent wrote, which the engine keeps so the link works.

More

The scanner, the plugin and four example apps are in the repository. MIT license.

Source: packages/connector/README.md at commit 8aa0ab4

Tools

0
SourceWeft has not started this package in its sandbox yet, so its tools are not known.

Version history

2
  1. v0.4.6LatestSep 26, 2026
  2. v0.4.5Sep 25, 2026

Score

50% of the applicable checks met

How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.

  • Reached in the last 30 days25 pts

    Keep the endpoint answering the MCP handshake; SourceWeft probes it every 30 days.

  • Auth declared10 pts

    Declare the credential headers in the registry entry, or keep the endpoint answering the probe so its auth can be classified.

  • Tools listed15 pts

    Answer tools/list without a credential, or ship a package that starts in a sandbox without one.

  • DescriptionRequired15 pts
  • Logo10 pts
  • Documentation15 pts
  • License10 pts