# lewm-mcp

> Use this tool when you need to detect visual anomalies in videos or UI changes, as it computes surprise scores between frames using a ViT encoder, enabling identification of unusual patterns. It takes in video frames as input and outputs surprise scores, indicating potential anomalies. Ideal for applications where monitoring visual consistency is crucial, such as quality control or surveillance.

Canonical page: https://skillsregistry.net/skills/gonzih-lewm-mcp  
JSON: https://api.skillsregistry.net/v1/skills/gonzih-lewm-mcp

## Description

Enables visual anomaly detection for Claude Code agents by computing surprise scores between frames using a ViT encoder, allowing detection of UI changes and video anomalies.

## Trust

- **Trust score (0–1):** 0.69
- **Verification tier:** scanned
- **Last scanned:** 2026-08-31

## Facts

- **Version:** 1.0.0
- **Skill type:** atomic
- **Execution layer:** mcp-remote
- **Runtime environment:** api
- **Category:** media
- **Updated:** 2026-08-31

## Source

- **Source listing:** [Glama](https://glama.ai/mcp/servers/dbgom3exf2)
- **Repository:** <https://github.com/Gonzih/lewm-mcp>

## Use it

Resolve this record through the SkillsRegistry MCP server (no auth, read-only):

```
claude mcp add --transport http --scope user skillsregistry https://api.skillsregistry.net/mcp
```

```json
{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "get_skill",
    "arguments": {
      "slug": "gonzih-lewm-mcp"
    }
  }
}
```

REST: `GET https://api.skillsregistry.net/v1/skills/gonzih-lewm-mcp` · pull for local use: `GET https://api.skillsregistry.net/v1/skills/gonzih-lewm-mcp/pull`

---
SkillsRegistry indexes agent skills from public registries and GitHub. Skills we have analysed are scanned with Circle-IR and scored on six dimensions; each listing states its scan coverage. More: https://skillsregistry.net/llms.txt