For site owners & operators
The Bramwell Browser Agent
If you're reading this, you probably found Bramwell-Agent in your server logs, or followed the contact link inside that User-Agent string. This page explains what made the request, on whose behalf, and how to reach us — including how to have your domain blocked for every Bramwell user.
What Bramwell is
Bramwell is personal-assistant software for Home Assistant, made by Bramwell Technologies LLC. It installs on a customer's own computer, in their home. Its browser agent drives a real browser on that machine when the customer asks for help with a task — working through a setup screen, checking a page, or running an errand they requested.
Each installation serves one household. There is no fleet, no crawler, and no central index: requests originate from the individual user's own internet connection, and what the agent reads is used only to complete that user's request.
What the agent is not:
- It is not a search-engine crawler and does not index or archive the web.
- It does not collect content for AI training.
- It does not act on its own initiative — it acts only at its user's direction, either live or on a schedule that user configured.
How it identifies itself
Bramwell identifies itself with two User-Agent strings. Browser sessions:
Bramwell-Agent/1.0 (+https://github.com/bengulliford/Bramwell)
Lightweight automated fetches (RSS feeds, structured-data lookups, geocoding):
Bramwell/1.0 (+https://github.com/bengulliford/bramwell)
Traffic carrying either string came from a Bramwell user's own installation, acting for that user.
How it behaves
- Automated fetches check robots.txt first. Scheduled, non-interactive fetches retrieve the site's
robots.txtbefore requesting a page and honorDisallowrules addressed to their user-agent token or to*. - They are rate-limited. At most one request per second per host, a 15-second timeout, and a small fixed cap on items read per check.
- Live sessions are human-scale. An interactive session is a single real browser, driven for a single household's request — not parallel crawling.
- Domains that ask to be left alone are left alone. The agent enforces a blocklist before navigation ever happens; blocked domains are refused for every user (see below).
Contact, opt-out, and blocking
- Open an issue on the public repository: github.com/bengulliford/Bramwell
- Or email [email protected]
When a site owner asks us not to visit their domain, we add it to a global blocklist that applies to all Bramwell users, enforced before the agent ever navigates. We honor these requests promptly, and we treat cease-and-desist notices the same way.
If something on this page doesn't match what you see in your logs, tell us — the same two channels reach the people who build this.