$ loading_
Detect UI changes and visual anomalies between frames for code agents.
Copy the install command and let the AI configure it · recommended for beginners
No copy-paste install info for "lewm-mcp" yet — see the docs or source repo.
Use lewm-mcp to compare frames from the screen recordings before and after this feature update, compute surprise scores, and identify the timestamps and regions with the largest abnormal UI changes with a brief explanation.
Returns frames with high anomaly scores, related timestamps, suspicious UI regions, and a brief assessment of possible regression issues.
Use lewm-mcp to analyze this automated test recording, find visually anomalous frames that differ from normal behavior, and rank them by severity.
Outputs a list of anomalous frames, ranked anomaly scores, and hints about the workflow issue each anomaly may indicate.
Use lewm-mcp to scan this operation video, detect sudden anomalous visual changes, and summarize the segments most worth manual review.
Provides key segments for review, the anomaly score trend, and a short explanation for each detected anomaly.
Search prompts, enhance them, and generate images or videos in MCP hosts.
Analyze videos with frame extraction, scene detection, and metadata retrieval.
Query Claude Code transcript analytics for cost, safety, audit, and efficiency insights.
Analyze images and videos with configurable model providers for visual understanding.
Add private local image and video frame understanding to AI assistants.
Give AI agents prompts, prompt enhancement, and image or video generation.