Senro

Senro is evaluation and observability infrastructure for WebMCP, the interface that lets AI agents discover and invoke website tools directly. Made by Senro Inc. and led by founder and CEO Nart Madi, it lets teams test their WebMCP tools before deployment and monitor agent usage in production.

What It Is

Senro addresses a reliability problem that comes with agent-facing web interfaces: agents may use a WebMCP tool successfully with one LLM or language and fail with another, select the wrong tool, pass incorrect arguments, or behave differently in production. Senro describes itself as the first dedicated evaluation and observability infrastructure for WebMCP, so teams do not have to build that reliability tooling themselves.

Evaluating Before Production

The pre-production side covers testing of tools, conversation, and outcome. The homepage lists smoke testing, goal evaluation, trajectory evaluation, page checks, simulated users, and multi-model comparison. Evaluation runs display model results and pass rates, which supports comparing how different models handle the same tools.

Observing in Production

In production, Senro monitors how agents use a site's tools. The homepage lists tool activity, inputs, results and errors, session timelines, production context, issue tracking, and usage and error metrics, presented through an observability dashboard of production activity and errors.