Feed URLs and compare captured data
Connect
SPIRITT
Browse AIBrowse AI
Web changes into action
SPIRITT runs your existing Browse AI robots, retrieves extraction results, and creates monitors for the sites that matter. Turn those results into maintained databases, decision tools, and workflows across your connected apps.
What you can do with Browse AI.
Fresh evidence
Run robots and create monitors to keep pricing, listings, and other captured website data current, with task history behind each update.
Batch work handled
Process URL lists through a suitable robot, collect task results, and separate failed extractions from usable records before downstream work begins.
Useful systems
Build sourcing portals, pricing dashboards, and research libraries around extracted fields, then route meaningful changes to the right team.
Works with your other apps.
Get started in three steps.
01
Open your workspace
Start a SPIRITT workspace for the website data and recurring research you want turned into useful work.
02
Connect Browse AI
Connect Browse AI with your API key so SPIRITT can access your robots, run tasks, and configure monitors for the mission.
03
Delegate an outcome
Name the sites, existing robots, and result you need. SPIRITT runs the extraction workflow and builds the system around its results.
Hand over a mission.
Choose a mission. Make it yours.
Put Browse AI to work.
Put your website data to work
Questions
Browse AI, with SPIRITT.
01Does SPIRITT create Browse AI robots or run the ones I already have?+
SPIRITT can list and run your existing robots, retrieve their task histories and detailed results, and create monitors and webhooks. Robot creation and training are not included in these actions, so the mission starts with a robot configured to capture the fields you need.
02What access do I need to connect Browse AI?+
Connect with a Browse AI API key for the account containing your robots. Give SPIRITT the relevant robot or extraction goal, the input URLs, and the destination for results. Connected destinations such as Google Sheets or Slack need their own access.
03Can SPIRITT process a large list of website inputs?+
Yes. Bulk runs can submit up to 1,000 tasks at a time to a specified robot. SPIRITT can split larger lists into batches, track the resulting tasks, collect their details, and separate unsuccessful extractions from records ready for downstream use.
04How do monitors and webhooks fit into recurring work?+
A monitor runs a robot to track website changes over time. A webhook sends callbacks to a receiving service. SPIRITT can create monitors, inspect existing robot webhooks, and create a webhook for a system it builds, then retrieve task details and process the results into connected tools.
05What can SPIRITT build from extracted website data?+
It can build pricing dashboards, supplier availability apps, tender queues, or searchable research libraries using fields captured by your robots. Results can feed Google Sheets, Airtable, Notion, or a Supabase-backed app, with task IDs and capture times retained so users can trace the underlying evidence.
06How is this different from just scheduling a Browse AI robot?+
A scheduled extraction produces website data. SPIRITT adds the operation around it: matching records, maintaining a custom interface, coordinating downstream tasks, and investigating failed runs using task details and system status. Browse AI handles the configured extraction while SPIRITT turns the results into completed business work.