Submit a resource

Navigation

Back to directory

Vibium

A browser automation toolkit for AI coding agents that lets them navigate pages, fill forms, click, and take screenshots to verify their own work.

Visit website (opens in a new tab)vibium.com
Vibium

Vibium is a verification layer that gives AI coding agents browser automation tools to check their own work. Installed as a skill, it lets an agent navigate pages, fill forms, click buttons, and take screenshots through simple CLI commands, and it is also available as an MCP server and as JavaScript/TypeScript, Python, and Java client libraries.

It is built on WebDriver BiDi rather than a proprietary automation protocol, ships as a single binary of about 10MB with no runtime dependencies, and downloads its own browser automatically on first setup. Beyond driving a browser directly, its check capability independently assesses whether a claim about a page or a recording actually holds, and it integrates with agents such as Codex, Claude Code, and Claude Desktop through slash commands or MCP.

It is written in Go and released under the Apache License 2.0.

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.

Cookies

We use necessary cookies to run the site. Measurement and marketing are optional; neither is used today. Privacy policy