mcpField
D
Demarker

Demarker

Local
demarker.ioMedia & Design0Updated 2w ago
RepositoryVisit website

About Demarker

Demarker is an MCP server for identifying and cleaning marks, metadata, and watermarks in AI images. Its identification tool works without a key, and cleaning saves results beside the input without overwriting it.

Key features

  1. 01Identify image marks and metadata
  2. 02Clean image content
  3. 03Check balance, pricing, and jobs
  4. 04Set a per-call cleaning credit cap
  5. 05Save results without overwriting inputs

Use cases

  1. 01Checking images for watermarks and marks
  2. 02Cleaning AI image content
  3. 03Checking cleaning jobs and account balance

MCP support

It exposes image identification, cleaning, balance and pricing lookup, and job lookup tools to MCP clients. Cleaning requires an API key, a credit limit, and user confirmation of the rights statement; identification does not require a key.

Configuration

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

mcp.json · Universal config
{
  "mcpServers": {
    "io-demarker": {
      "command": "npx",
      "args": [
        "-y",
        "@demarker/mcp"
      ],
      "env": {
        "DEMARKER_API_KEY": "<DEMARKER_API_KEY>",
        "DEMARKER_MCP_MAX_CREDITS": "<DEMARKER_MCP_MAX_CREDITS>"
      }
    }
  }
}
Run commandnpx -y @demarker/mcp
Environment variables to set
DEMARKER_API_KEYSecretOptional. Needed to clean images and read balance. Create one at https://demarker.io/app/settings/api
DEMARKER_MCP_MAX_CREDITSOptional. Hard ceiling on credits one clean call may spend. Default 2.

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
npm@demarker/mcp0.1.2

Tools

5

Saved tool list · updated 2w ago

ParameterTypeRequiredDescription
pathstringYesAbsolute path to a PNG, JPG, WebP or AVIF file.

Details

From the project README

@demarker/mcp

MCP server for Demarker: identify what an AI image carries, then clean it.

Install

  • Claude Code: claude mcp add demarker -- npx -y @demarker/mcp
  • Cursor, VS Code: one-click buttons at https://demarker.io/developers
  • Claude Desktop, Windsurf, Cline:
{ "mcpServers": { "demarker": { "command": "npx", "args": ["-y", "@demarker/mcp"],
  "env": { "DEMARKER_API_KEY": "sk_live_YOUR_KEY" } } } }

Tools: demarker_identify (free; works without a key), demarker_clean (spends credits), demarker_get_balance, demarker_get_pricing, demarker_get_job.

Safety

  • demarker_clean needs maxCredits, and a hard per-call ceiling applies whatever the model asks for (default 2, set DEMARKER_MCP_MAX_CREDITS).
  • The user must confirm the rights statement. Clients that support elicitation show it to the user; others must pass rightsAttested: true only after the user explicitly agreed.
  • Results are written next to the input and never overwrite a file.

Env: DEMARKER_API_KEY (to clean; create at https://demarker.io/app/settings/api), DEMARKER_API_URL, DEMARKER_MCP_MAX_CREDITS.

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.