UI Debugger MCP
AI coding agents (Claude, etc.) are great at writing code. They're bad at knowing if the UI actually works. For backend code there are unit and integration tests. For UI, a human still has to open the app, log in, click around, and report what's broken. That human-in-the-loop is slow, boring, and the main bottleneck when an entire product is built by AI.
People who work with browser automation, cdp and openrouter and want it reachable from Claude, Cursor, VS Code, or another MCP client. The project is written in TypeScript.
VERIFIED ACTIVE
LAST COMMIT 2026-09-17 · ★ 2 · #104 OF 255 MAINTAINED BROWSER AUTOMATION · VERIFIED 2026-09-18
MIT · TypeScript servers · how we verify → /methodology
01 · Install UI Debugger MCP
before you install - you'll need
Set OPENAI_API_KEY, OPENAI_BASE_URL before connecting.
Claude Code
claude mcp add developerz-ai-ui-debugger-mcp -- npx -y @developerz.ai/ui-debugger-mcp Claude Desktop / Cursor / VS Code - add to config
{
"mcpServers": {
"developerz-ai-ui-debugger-mcp": {
"command": "npx",
"args": [
"-y",
"@developerz.ai/ui-debugger-mcp"
]
}
}
} Same JSON for Cursor. For VS Code, rename the top-level key from `mcpServers` to `servers`.
Using another client? Same JSON, different key
Claude Desktop · mcpServers
Cursor · mcpServers
VS Code · servers
Windsurf · mcpServers
Zed · context_servers
Cline · mcpServers
Roo Code · mcpServers
Continue · mcpServers
LibreChat · mcpServers
Gemini CLI · mcpServers
Codex CLI · mcp_servers
Full setup guides: every client.
02 · Evidence
Security posture
What to check before giving this server access to your agent - from the registry, GitHub, and our own probes. We don't score safety; we show what's verifiable.
runs as local process (stdio) - runs on your machine with your user's permissions
license MIT - declared in the repository
npm package @developerz.ai/ui-debugger-mcp - the @developerz.ai scope differs from the repo owner (developerz-ai); confirm it's the project's official package
registry namespace io.github.developerz-ai is GitHub-verified and matches the repo owner
03 · What UI Debugger MCP can do
Prose above is summarized from the project's README and registry record - no invented capabilities.
Latest releases
v1.9.0 · 2026-09-05
Added · The per-repo config is read from .dz/ui-debugger/ui-debugger-mcp.json first, falling back to the root .ui-debugger-mcp.json. A repo that has consolidated its agent config under .dz/ now boots the tool instead…
v1.8.0 · 2026-07-31
[1.8.0] - 2026-07-30 · Changed · A managed android run is headless by default. The adapter used to spawn · emulator @avd -port <p> and nothing else, which opens a window on a desktop · and fails to boot at all on a box…
v1.7.0 · 2026-07-27
[1.7.0] - 2026-07-27 · Two ways a run went wrong before it started: the project was still held by a run · nobody could name any more, or the driver had never been told that this app's · empty tables are the point.…
04 · Who maintains UI Debugger MCP
ui-debugger-mcp is maintained by developerz-ai. It's the only MCP server we track from this author; the repo dates to Jun 2026.
05 · Facts
- repository
- github.com/developerz-ai/ui-debugger-mcp
- category
- browser automation - ranked #104 of 255 actively-maintained browser automation servers as of 2026-09-18.
- release cadence
- 10+ releases in the last 90 days (latest 2026-09-05)
- registry
- io.github.developerz-ai/ui-debugger-mcp (active, first published 2026-07-24 · 10 versions)
- packages
- npm:@developerz.ai/ui-debugger-mcp
06 · UI Debugger MCP FAQ
What is UI Debugger MCP?
AI coding agents (Claude, etc.) are great at writing code. They're bad at knowing if the UI actually works. For backend code there are unit and integration tests. For UI, a human still has to open the app, log in, click around, and report what's broken. That human-in-the-loop is slow, boring, and the main bottleneck when an entire product is built by AI.
Is UI Debugger MCP still maintained?
Yes - as of 2026-09-18, its last commit was 2026-09-17 and it shipped 10+ releases in the last 90 days. We re-verify nightly.
How do I install UI Debugger MCP?
Run `npx -y @developerz.ai/ui-debugger-mcp`. The README documents 2 environment variables (OPENAI_API_KEY, OPENAI_BASE_URL) to set first. Set OPENAI_API_KEY, OPENAI_BASE_URL before connecting. You can also paste the ready-made client config above.
Does UI Debugger MCP run locally?
Yes - it's a stdio server: it runs on your machine (via npx) with your user's permissions. Your data stays local unless the server itself calls external APIs.
07 · Alternatives to UI Debugger MCP
Alternatives to UI Debugger MCP
Maintained browser automation servers if UI Debugger MCP isn't the fit.
- Chrome DevTools MCP MCP server for Chrome DevTools ★ 52,170 · 2026-09-17
- Server Browser MCP server for browser use access ★ 39,038 · 2026-09-11
- Playwright MCP Playwright Tools for MCP ★ 37,231 · 2026-09-17
- Aihawk AI browser agent: browses, clicks, types, and reads real web pages from plain-English instructions. ★ 31,628 · 2026-09-18
- SeleniumBase MCP Stealthy browser automation, testing, and web-scraping via CDP Mode. ★ 13,024 · 2026-09-18
- Mobile MCP MCP server for iOS and Android Mobile Development, Automation and Testing ★ 6,744 · 2026-09-16
Pairs well with
Servers that cover what UI Debugger MCP doesn't - only shown when the pairing reason fits the companion.
- Firecrawl MCP Server → web scraping companion web scraping · ★ 7,478
- Tavily MCP → web scraping companion web scraping · ★ 2,390
- Codebase Memory → memory companion memory · ★ 43,715
- Memorix → memory companion memory · ★ 785
- Server Filesystem → filesystem companion filesystem · ★ 39,038
- Pdf MCP → filesystem companion filesystem · ★ 135
More browser automation MCP servers · Swell MCP · Dgmo MCP · Google data (Trends, YouTube, Ads Transparency, Play) · Thinkrun · E2 LLM
More TypeScript MCP servers · Dev Lens · Gitwand · x402 Trinity · SendGrid MCP Server · see all