VisionSqueezer MCP
VisionSqueezer is an LLM-native image optimization MCP server that shrinks images to fit a vision-LLM token budget before they reach the model. It supports various clients like Claude Code, Cursor, and Codex, helping agents and pipelines reduce costs by optimizing images sent to model APIs or read from files.
Package and access
Local package with source-attributed installation metadata. Koskamo does not execute packages during review.
What you can do with it
- Shrink images to fit vision-LLM token budgets
- Optimize images for agents and pipelines
- Reduce API costs for vision models
Source and maintenance
- Maintenance checked
Packages and installation
Source-provided package data. Koskamo has not executed or live-probed this package.
vision-squeezer 0.9.0
{
"version": "0.9.0",
"transport": "stdio",
"identifier": "vision-squeezer",
"registryType": "npm"
}Tools
Reported by the attributed source; not independently observed by Koskamo.
Connect from your client
Each snippet configures the same source-provided package; replace placeholders before use.
Cursor
{
"mcpServers": {
"visionsqueezer-mcp": {
"command": "npx",
"args": [
"-y",
"vision-squeezer@0.9.0"
]
}
}
}Cline
{
"mcpServers": {
"visionsqueezer-mcp": {
"command": "npx",
"args": [
"-y",
"vision-squeezer@0.9.0"
]
}
}
}VS Code
{
"servers": {
"visionsqueezer-mcp": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"vision-squeezer@0.9.0"
]
}
}
}Claude Desktop
{
"mcpServers": {
"visionsqueezer-mcp": {
"command": "npx",
"args": [
"-y",
"vision-squeezer@0.9.0"
]
}
}
}Let your agent find its tools
Find servers like VisionSqueezer MCP from your agent. Add Koskamo once to search the index for endpoints, access requirements, and client setup. Public discovery needs no key.