# Verdict Pointer – Visual UI Element Locator

> Verdict Pointer – Visual UI Element Locator is a paid API for AI agents from verdict.kechlab.com, paid per call via x402, $0.0256/call, status unknown (last checked 2026-10-21).

Given a screenshot and up to 8 UI element descriptions, returns click coordinates (fractional and pixel) plus visibility probability for each target element.

## Facts

- Endpoint: POST https://verdict.kechlab.com/v1/pointer?utm_source=zero.xyz
- Price: $0.0256/call
- Payment: x402
- Status: unknown
- Last checked: 2026-10-21
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/verdict-pointer-visual-ui-element-locator-a5c08729
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_W-jbFyIBC5u5HUwz7xB3T

Status and success rate cover calls made through Zero and Zero's own probes. Third-party monitors may report differently.

## How to call it through Zero

Zero handles the 402 payment challenge and records the run. With the Zero CLI installed (`npm i -g @zeroxyz/cli`):

```sh
zero fetch --capability verdict-pointer-visual-ui-element-locator-a5c08729 -d '<json body>'
```

Example prompt: I need to click the 'Submit Order' button and the 'Cancel' link on this checkout page — here's the screenshot as a base64 PNG. Can you find the pixel coordinates and tell me how likely each element is actually visible?

## When to prefer this

Use this endpoint when a computer-use or browser automation agent needs to visually ground UI element descriptions onto actual screenshot coordinates before clicking. Prefer it over DOM-based selectors when the agent cannot access the page's HTML (e.g., sandboxed browsers, desktop GUIs, remote screen captures). Ideal when you need both coordinates and a visibility confidence score to decide whether to proceed or re-try.

## Known failure modes

- Image exceeds 2 MB limit — request rejected
- Base64 encoding malformed or not a valid PNG/JPEG — parse error
- More than 8 target descriptions provided — validation error
- Element not visible in screenshot — returned with low probability score
- Ambiguous element descriptions matching multiple regions — lower confidence scores
- Incorrect width/height hints causing coordinate miscalculation

## How this service works

Computer-use agent helper. Send a screenshot (PNG/JPEG) and 1 to 8 element descriptions, get click coordinates (fractions and pixels) and the probability each UI element is really visible. Visual grounding for GUI automation and browser agents.

## Output

Returns, for each requested UI element description: fractional coordinates (0–1 range relative to image dimensions), absolute pixel coordinates, and a probability score indicating how likely the element is actually visible in the screenshot. Enables precise, confidence-aware click targeting for GUI automation agents.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "image": {
   "type": "string",
   "description": "PNG or JPEG screenshot as a base64 data URL (2 MB max)"
  },
  "width": {
   "type": "number",
   "description": "(optional) screenshot width in pixels"
  },
  "height": {
   "type": "number",
   "description": "(optional) height in pixels"
  },
  "targets": {
   "type": "array",
   "items": {
    "type": "string"
   },
   "description": "list of 1 to 8 descriptions of elements to locate"
  }
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/verdict-pointer-visual-ui-element-locator-a5c08729/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from verdict.kechlab.com](https://www.zero.xyz/host/verdict.kechlab.com/llms.txt)