bots.st/captureoneai-web-perception
CaptureOneAI Web Perception
Page context, element extraction, change detection
2005 ms
answered our call
- State
- Answered
- Protocol
- MCP 2025-06-18
- Tools listed
- 6
- Answered
- 3 of the last 3 calls
How to reach it
- MCP, HTTP
https://captureoneai.com/mcp
What it can do
| Tool | What it says it does |
|---|---|
capture_context | Returns the structured representation of a web page: visible text with coordinates, interactive elements, forms and tables. Use this to let an agent read and understand a page without parsing raw HTML. |
check_monitor | Triggers an immediate check for a monitor instead of waiting for its next scheduled run, and returns any detected changes. |
create_semantic_monitor | Watches a URL on a schedule and reports what changed and whether it matters, explained by AI — not just a pixel-diff percentage. Use this for competitor pricing pages, changelogs, status pages, or any page whose changes should trigger a workflow. |
find_elements | Returns only the interactive elements of a page (buttons, links, inputs) with role, label, CSS selector and pixel coordinates. Use this before acting on a page (computer-use / Set-of-Marks style grounding) instead of capture_context when you only need to know what can be clicked and where. |
get_monitor_changes | Returns the recent check history for a monitor, including detected changes with their relevance and explanation. |
list_monitors | Lists your active monitors and their IDs, URLs and schedules. |
In its maker's words
Give agents eyes on any web page: structured context, and changes explained in plain language.
- Website
- captureoneai.com
- Source
- gitlab.com/zaisor-technology/capture-one-ia
- Publisher
- captureoneai.com
- Listed from
- the official MCP Registry as
com.captureoneai/web-perception - Version
- 1.0.0, updated 21 Sep 2026
- Kinds of work
- Browser and automation, AI models and agent tools, Search and research
The domain it lives on
- captureoneai.com
- registered 1 May 2026 (6 months ago), at NameCheap, Inc., its home page is a working site
From our own crawl of domain names: registry records and the home page as our crawler last saw it.
Every call we made
| When | What happened | Time |
|---|---|---|
| Answered | 2005 ms | |
| Answered | 2210 ms | |
| Answered | 421 ms |
Next door
- Bowmark web data: LinkedIn, Amazon, YouTube, Reddit, Airbnb, InstagramScrape and automate logged-in web data
- SkyelightPin feedback to rendered elements with code
- TapAutomate and validate logged-in browser tasks
- String Web AccessFetch URLs, crawl sites, search web, return clean Markdown
- ShipvoSimulate a phone and automate UI actions
Is this your bot? Claim its address to add a verified badge and keep the listing current.