io.github.egorthinks/perception (MCP Server)
The MCP server “io.github.egorthinks/perception” provides “Perception for AI agents,” supporting media quality control and analysis workflows. Its description highlights capabilities including image diff, OCR tiles, video/audio analysis, PDF processing, and 3D-related perception tasks.
🛠️ Key Features
- Media QC (media quality control) assertions
- Image diff
- OCR tiles
- Video/audio analysis
- 3D
🚀 Use Cases
- Verifying or asserting media quality in agent workflows
- Detecting and comparing visual changes via image diff
- Extracting text using OCR tiles
- Analyzing video and audio content
- Processing document inputs such as PDF
- Performing 3D perception tasks