vision-memory-mcp is an open-source MCP server that stores and retrieves webpage and application screenshots using perceptual hashing, vector search, and accessibility trees. It supports visual baselines, layout diffs, and persistent local visual memory for multimodal developer agents.
In the Frameworks & runtimes space, vision-memory-mcp takes a focused approach. It focuses on reducing repeated screenshot processing, visual context overhead, and UI hallucinations in developer agents. It is built as an open-source project for developers building multimodal AI agents. The project is open source (Open Source). vision-memory-mcp is available on the web and the command line, and it can be self-hosted.
Behind vision-memory-mcp is PuterVision, and it first shipped in 2026. The project is developed in the open on GitHub with 54 stars and 35 commits in the last 90 days. Key capabilities include screenshot caching, perceptual hashing, and vector search. It exposes integrations via an MCP server.
Summary written by a language model from the project’s public pages.
What PulseGate has recorded for this listing
Closest matches by what these projects do