mcpField
O
optiq-mcp

optiq-mcp

Local
github.com/BaukaalmOther0Updated today
Repository

About optiq-mcp

optiq-mcp is a local MCP server for optimizing project images without uploading them. It verifies output quality using SSIM thresholds or pixel-exact checks before replacing files.

Key features

  1. 01Analyze images in a dry run
  2. 02Optimize JPEG, PNG, WebP, and AVIF
  3. 03Lossless or SSIM-based quality checks
  4. 04Write in place or to an output directory
  5. 05Resize images and strip metadata

Use cases

  1. 01Preview image size savings before writing
  2. 02Reduce project image file sizes
  3. 03Preserve image paths with in-place optimization
  4. 04Remove EXIF and XMP metadata

MCP support

The server runs locally over stdio and exposes image analysis and optimization tools to MCP clients.

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

mcp.json · Universal config
{
  "mcpServers": {
    "baukaalm-optiq": {
      "command": "npx",
      "args": [
        "-y",
        "optiq-mcp"
      ]
    }
  }
}
Run commandnpx -y optiq-mcp

Local servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.

Packages
npmoptiq-mcp0.1.0-beta.0

Tools

No tool list yet

Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.

Details

From the project README

optiq-mcp

An MCP server that lets AI agents optimise the images in a project with Optiq. It runs locally over stdio, and nothing is uploaded.

claude mcp add optiq -- npx -y optiq-mcp          # Claude Code

{ "mcpServers": { "optiq": { "command": "npx", "args": ["-y", "optiq-mcp"] } } }

The JSON form is for Claude Desktop, Cursor and other clients. For VS Code (.vscode/mcp.json), use { "servers": { "optiq": { "type": "stdio", "command": "npx", "args": ["-y", "optiq-mcp"] } } }.

Tools

ToolWritesWhat it does
analyze_imagesno (readOnlyHint)Dry run. For each image it reports the chosen format, the bytes before and after, and the measured quality (SSIM, or pixel-exact)
optimize_imagesyeswrite: "in-place" keeps formats and names, so references in code stay valid. write: "out-dir" may change formats

What each mode guarantees:

  • lossless: pixels are verified identical;
  • visually-lossless: SSIM ≥ 0.99;
  • balanced (the default): SSIM ≥ 0.975;
  • fast: the same threshold as balanced, but no AVIF and fewer encodes.

A file is replaced only when the result passes the check and is smaller. Lossy outputs carry a ~20-byte marker, so running optimize_images again does not re-encode them and cannot compound the loss (reoptimize: true overrides this).

Other options:

  • mode;
  • keepFormat;
  • formats;
  • maxSize (downscale the longest side);
  • stripMetadata (removes EXIF including GPS, and XMP);
  • minQuality;
  • minSaving.

Full docs: https://baukaalm.github.io/optiq/guide/agents

Requires Node ≥ 20. Apache-2.0. The codec licences are listed in the optiq-image NOTICE.

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.