You read the exact markdown
The panel shows it as plain text, never rendered, alongside the domain it will match. Activate arms it, Discard deletes it.
Skills
Skills:
automatic
site
maps,
recorded
jobs
and
instant
commands
Point the agent at a web app and it maps the site. Record a job once and it keeps the steps. Both become skills you review and activate yourself.
Automated site discovery
Map a web app automatically in a few minutes
An agent new to a site spends its first minutes finding its way around. A site map does that once: the agent reads the site’s robots.txt and sitemap.xml, then explores for a few minutes, taking screenshots as it goes. Every later instruction on that domain starts from the map.
Invoke with
@site-mapper map this site
-
Nothing takes effect until you say so
A map in progress is written to a staging directory the skill loader cannot read, so an unreviewed map is never opened.
-
You read the exact markdown
The panel shows it as plain text, never rendered, alongside the domain it will match. Activate arms it, Discard deletes it.
-
Read-only and locked to one host
It cannot click, fill or submit, cannot leave the site, and stays pinned to the tab it started in.
-
Ceilings you can narrow, never widen
A short page budget and a short clock by default, under hard limits Browsentic Bridge enforces above them.
Crawl & map
- ~/browsentic/skills/acme-com/
- ├── SKILL.mdlandmarks, key pages, quirks
- ├── map.jsonthe structured report behind it
- ├── screenshots/captures taken during the crawl
- ├── evidence/the robots.txt and sitemap it worked from
- └── pages/per-page notes, kept out of the prompt
Record once, replay later
Record a job once, then replay it with one instruction
Press record, do the job yourself across as many pages as it takes, then press stop. Browsentic turns what you did into ordered steps, each named after what it achieved. From then on, “do it like last time” is the whole instruction.
Invoke with
record my browsing session
-
What you type is not saved by default
Every field you fill becomes a placeholder ({{email}}, {{invoice_number}}) and the assistant asks you for the value when it replays.
-
Some things are never stored either way
Passwords, hidden fields, one-time codes and anything shaped like a card number are dropped whether or not you opt in.
-
Replay checks each step against the live page
The agent re-checks each target before acting and prefers the visible text it recorded over the CSS selector.
-
A step that fails stops the run
If a step no longer lands, the run halts and tells you which one, instead of finding a different route to the same effect.
Record & replay
- Recordingstops itself at the ceiling
- ✓ Step 1Open the invoices list
- ✓ Step 2Filter to unpaid
- ✓ Step 3Fill {{invoice_number}}
- ● Step 4Submit (waits for your approval)
Local intent classification
Simple commands run locally, without the model
Every instruction is first scored against a local grammar. Confident single-step commands, such as “go back”, run in the browser and stop there. Everything else goes to the agent unchanged. When unsure it escalates, because a misread command would spend a wrong click on your real page.
Invoke with
yarn check:intent "take me to the checkout page"
-
Runs locally, in milliseconds
back, forward, reload · open github.com · scroll to the top · press enter · click Sign in · google something · stop recording
-
Goes to the agent
“is there a login button?” · “open the settings menu” · “click Buy now” · “scroll down and tell me what it says” · “click it”
-
Questions always escalate
So do multi-step asks, conditionals, vague targets and consequential clicks, as does any local command that runs and fails.
-
You can audit a single decision
Route one utterance through the grammar from the command line and see exactly why it went where it went.
Intent routing
- ⚡ go backlocal
- ⚡ scroll to the toplocal
- → is there a login button?agent
- ⚡ open github.comlocal
- → click Buy nowagent · consequential
Review
A skill does nothing until you review and activate it
A map in progress goes to a staging directory the skill loader cannot read, so an unreviewed map is never opened.
Locked to one host
A mapping run is read-only. It cannot click, fill or submit, cannot leave the site, and stays pinned to the tab it started in.
Ceilings you can narrow, never widen
A short page budget and a short clock by default, under hard limits Browsentic Bridge enforces above them.
Give your browser a superpower.
Free and open source under Apache 2.0, with no API key, subscription or account. One line installs Browsentic Bridge and opens the extension’s store page. Nothing connects until you pair the two.
Available in the Chrome Web Store
Available on Edge Add-ons
Firefox add-on, signed by Mozilla