Back to directory

kogiQA MCP Web Browser

An MCP server that gives coding agents a browser they can drive without CSS selectors, for debugging style issues, fixing console errors, mapping a web app's pages, and writing end-to-end tests.

kogiQA MCP Web Browser

This MCP server lets an LLM interact with web pages through natural-language instructions instead of screenshots or visually tuned models, using the kogiQA browser-control algorithm so an agent can act on a page without first finding a CSS selector, which the project says saves both time and tokens compared with a selector-based approach such as Playwright MCP.

Suggested uses include debugging style issues on a page, automatically fixing console errors, mapping and documenting a web app's functionality, writing end-to-end tests, and automating exploratory testing of an application. It installs with a single npx command and is configured into MCP clients such as Claude Code, VS Code or Cursor.

More in Testing

See all 69
  • Canary

    runcanary.ai

    An independent testing service, backed by Y Combinator, that runs an application and tries to break every change an AI coding agent makes, reporting what would have reached production.

  • MoEWatch

    github.com

    A diagnostic and audit library for Mixture-of-Experts models in HuggingFace Transformers that instruments router modules to catch expert collapse, routing entropy collapse and load imbalance during training.

  • Mockwave

    github.com

    An open-source, multi-protocol mock server for HTTP, GraphQL, SOAP, and gRPC, with weighted traffic splitting, dynamic JavaScript responses, chaos-testing fault injection, and an MCP server so an AI assistant can generate mock rules directly.