Sitelooper MCP server

sitelooper

Write browser tests in plain English. Run them in CI as plain Playwright. Repair them automatically when the app changes.

sitelooper is a CLI with a small LLM agent inside. You give it one instruction at a time, such as "sign in as ops@example.com, create a ticket titled 'k7 Bench' and report its id". It works the live browser and returns one verified result. Every step it gets right is compiled into a procedure that replays with no model, no tokens and no cost, and can be exported as a standalone @playwright/test spec.

Recorded once against the in-repo repair-desk app (24s, 28s and 42s with the agent driving). The app is reset, the same flow replays in 16s with zero model calls, and then the app's own state is queried.

How it works

locators, where each value came from, and what each step changed on the page.

1. Record. A cheap model follows your instructions. As it works, sitelooper records durable

that step goes back to the model, and a recovery that validates is pinned into the flow, so the flow heals itself over runs.

1. Replay. The recording replays deterministically. If the app has changed under a step, only

runs in CI with no sitelooper daemon and no model.

1. Compile. sitelooper build emits a Playwright spec with effect checks on every step. It

Results

Ten self-hosted apps, each job done once to record it, then twice more with no model, then as a compiled Playwright spec. The comparator is agent-browser driven by an LLM, which redoes the job on every run. Every run was scored against the app's own database or API, never against the tool's own report. September 2026:

instructions and DeepSeek V4.1 Flash driving the browser. Every replay ran with zero model calls. Nine of the ten compiled specs pass; EspoCRM's refuses to compile rather than guess.

• sitelooper verified every objective in all 30 runs, with GPT-6 Luna writing the

EspoCRM and Snipe-IT, posted duplicate comments on three apps, and twice reported work as done that the app had not saved.

• agent-browser on the same model was clean on 3 of 11 runs. It failed outright on Odoo,

pays that price on every run, and on Grafana it again reported unsaved settings as verified.

• On GPT-6.1 Sol, about 20× the per-token price, agent-browser was clean on 9 of 10. It still

Report-only objectives ask a run to state a value in its final report, which a compiled spec does not write.

Shortened. The full README is on GitHub.