Skip to main content
Glama

How to connect SmugMug to Claude Code, ChatGPT and any AI agent

Yes, Claude, ChatGPT and other AI agents can work in your SmugMug account. SmugMug's API v2 needs an API key, OAuth and a developer. Cotlet.dev connects your own signed-in SmugMug account to any AI agent with no code, so you manage galleries in plain words.

What you can ask

Create a gallery "Smith Wedding 2026", make it private and share the link with the client.

Which galleries got no visits in the last 6 months?

Rename the photos in my latest gallery by date and location.

Set the price list for the new portrait galleries to my standard one.

List every gallery that is still public but marked client-only in its title.

Related MCP server: Google Photos MCP

Connect in 2 minutes

Get your API key at https://cotlet.dev/app and export it as COTLET_API_KEY.

Claude Code

claude mcp add --transport http cotlet https://cotlet.dev/mcp/v1 --header "Authorization: Bearer $COTLET_API_KEY"

ChatGPT

Settings → Apps → add connector URL https://cotlet.dev/mcp/v1, then sign in with Cotlet.dev. No API key needed.

Cursor

Copy .cursor/mcp.json from this repo into your project:

{
  "mcpServers": {
    "cotlet": {
      "url": "https://cotlet.dev/mcp/v1",
      "headers": {
        "Authorization": "Bearer ${COTLET_API_KEY}"
      }
    }
  }
}

Run it locally (stdio)

This repo is a real MCP server. It speaks MCP over stdio, lists its SmugMug tools without any key, and forwards every tool call to the hosted Cotlet.dev server. The hosted URL above stays the recommended way. Use this when your client only starts local servers.

{"mcpServers":{"smugmug":{"command":"uvx","args":["smugmug-mcp"],"env":{"COTLET_API_KEY":"your-key"}}}}

Or with Docker:

docker build -t smugmug-mcp .
docker run -i --rm -e COTLET_API_KEY smugmug-mcp

Tools: smugmug_guide, smugmug_status, smugmug_open, smugmug_read, smugmug_click, smugmug_type, smugmug_close. Without COTLET_API_KEY a tool call answers: "Set COTLET_API_KEY. Get a key at https://cotlet.dev/app".

Command line

Python 3.10+, nothing else to install. From PyPI:

pip install smugmug-mcp        # or run it once: uvx smugmug-mcp connect
smugmug-mcp connect

Or straight from this repo:

export COTLET_API_KEY="your-key"
python3 smugmug_agent.py connect   # ready config for Claude Code, Cursor, ChatGPT
python3 smugmug_agent.py status    # is SmugMug ready in your Cotlet.dev account
python3 smugmug_agent.py guide     # SmugMug screens and sign-in link
python3 smugmug_agent.py read      # open SmugMug, read the page, close the browser
python3 smugmug_agent.py tools     # list Cotlet.dev tools
python3 smugmug_agent.py serve     # start the stdio MCP server (same as no argument)

Questions from SmugMug users

Does SmugMug have an API?

Yes, SmugMug has API v2, but it needs an API key, OAuth and code. Cotlet.dev gives your AI agent the same work in your own SmugMug account with no API key and no developer.

How do I connect SmugMug to ChatGPT?

Add Cotlet.dev as an app in ChatGPT (Settings → Apps, connector URL https://cotlet.dev/mcp/v1), sign in to SmugMug once in your Cotlet.dev Cloud Browser, and ask ChatGPT in plain words. No code and no SmugMug API key.

Is there a way to use Claude with SmugMug?

Yes. In Claude Code run the one-line claude mcp add command above; in the Claude app add https://cotlet.dev/mcp/v1 as a custom connector. Claude then works in your own SmugMug account.

Is there a SmugMug MCP server?

SmugMug has no official MCP server. Cotlet.dev is a hosted MCP server that works in SmugMug for you, 24/7, with Claude, ChatGPT, Cursor, Codex, n8n, Make and Zapier.

Find it on

Also works with


More: https://cotlet.dev/integrations/smugmug-api-mcp-server?utm_source=github&utm_medium=readme&utm_campaign=smugmug

Start free: https://cotlet.dev/signup?app=smugmug&utm_source=github&utm_medium=readme&utm_campaign=smugmug

Available Tools

7 tools
smugmug_clickBInspect

Click a numbered element on the current SmugMug page.

Inputs: element (required).

smugmug_closeAInspect

Close the SmugMug browser session when the work is done, so credits stop running.

Inputs: no named parameters.

smugmug_guideAInspect

How to work in SmugMug (galleries and photo uploads): its screens and the sign-in link. Call this first.

Inputs: task.

smugmug_openBInspect

Open a SmugMug page in your signed-in Cloud Browser (default https://www.smugmug.com) to work with galleries and photo uploads.

Inputs: url.

smugmug_readAInspect

Read the current SmugMug page: its text and numbered buttons, links and fields.

Inputs: element.

smugmug_statusAInspect

Check that your Cotlet.dev account and SmugMug are ready: account, credits and SmugMug connection.

Inputs: no named parameters.

smugmug_typeBInspect

Type text into a numbered field on the current SmugMug page.

Inputs: text (required), element (required).

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    A3.7/5.0

    Scored across 7 tools

    Disambiguation5/5

    Each tool maps to a distinct step in a browser-automation workflow: guide (instructions), status (account readiness), open, read, click, type, close. The open/read/click/type/close primitives are classic and non-overlapping, and guide vs. status are clearly separated as onboarding instructions vs. live account checks.

    Naming Consistency4/5

    All tools share the smugmug_ prefix and snake_case, so the namespace is predictable. Minor deviation: most names are verbs (open, read, click, close, type) but guide and status are nouns, slightly breaking the verb_noun pattern.

    Tool Count5/5

    Seven tools is well-scoped for a site-specific browser automation server; each tool (guide, status, open, read, click, type, close) has a clear, necessary role in the workflow.

    Completeness4/5

    The core automation lifecycle (open, inspect, interact, close) is fully covered, with a guide and status check as helpful extras. Gaps exist around photo uploads specifically (no explicit file-selection/upload tool) and secondary interactions like scrolling or dropdown selection, though agents may work around some via type/click.

    Maintenance

    ActivityMaintained
    ResponsivenessNo recent community issues

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      A
      quality
      B
      maintenance
      Enables AI agents to work with a user's Google Photos through Google's picker, letting users choose exactly what the agent sees while providing tools for uploading media, creating and sharing albums, writing captions and enrichments, searching app-created media, and managing multiple Google accounts.
      29
      29 npm
      1
      MIT
    • A
      license
      Not graded
      quality
      B
      maintenance
      Enables AI coding agents to create client-editable photo and video galleries, upload and organize media in bulk, generate embed code for popular site builders, and manage client feedback — so non-technical clients can update their own content without a CMS.
      70 npm
      1
      MIT