# refreshdotdev/web-eval-agent

An MCP server that autonomously evaluates web applications.

Repository: https://github.com/refreshdotdev/web-eval-agent
Canonical: https://ross.abutalabs.com/products/web-eval-agent
Homepage: https://www.operative.sh/mcp
Language: Python
License: Apache-2.0
License Family: permissive
Topics: debugging, debugging-tool, mcp, mcp-server, modelcontextprotocol, vibe-coding, vibe-testing, playwright, qa
Archived: true
Last push: 2026-02-11T14:55:42+00:00

## Health v2 (maintenance only)
Score: 10/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 67, release rhythm 35, longevity 37
- inputs: {"age_days": 519, "days_push": 203, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, archived
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1240, forks 108 (observed 2026-08-28T04:04:06.208296+00:00)

## What it is
An MCP server that launches a browser-use powered agent to autonomously test and debug web applications from within AI code editors like Cursor. It captures screenshots, console errors, and network traffic, returning a UX report to the coding agent. The project has been officially sunset/discontinued.

## Use cases
- autonomously test my web app end-to-end from cursor
- let my coding agent verify its own changes in the browser
- capture console errors and network traffic during automated web testing
- run a signup flow and report UX issues automatically
- set up browser login state for repeated automated QA runs

## When to choose
- you use Cursor, Cline, or Windsurf and want autonomous browser-based QA of web apps
- you want an MCP tool that feeds console and network logs back into your coding agent's context

## When to avoid
- you need a maintained or supported tool - the project has been discontinued
- you want fully local operation without an operative.sh API key
- you need non-web (mobile/desktop) app testing

## Facets
- artifact type: service
- maturity: abandoned
- function: mcp, testing, e2e-testing, agent-framework
- domain: developer-tools, testing, web-development
- platform: python
- tags: mcp-server, browser-automation, playwright, qa, vibe-coding, autonomous-debugging, sunset, ai-agents, macos, linux, nodejs

## Member repositories
- refreshdotdev/web-eval-agent (main) score 10

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:06.208296+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T05:09:54.524236+00:00, confidence not recorded.
  - readme: https://github.com/refreshdotdev/web-eval-agent (fetched 2026-08-28T04:04:06.208296+00:00, sha 76d7392abd7b)
  - homepage: https://www.operative.sh/mcp (fetched 2026-08-29T12:20:54.976502+00:00, sha f89719377180)
- Data as of 2026-08-30T08:39:29.467469+00:00.