Skip to main content
Glama

Server Configuration

Describes the environment variables required to run the server.

NameRequiredDescriptionDefault
CODEX_HOMENoCodex home directory. Defaults to ~/.codex.
CODEX_PATHNoPath to the codex executable. Auto-detected if not set.
CODEX_MODELNoModel that orchestrates the image_gen call. Defaults to the Codex default.
CHATGPT_IMAGE_OUTPUT_DIRNoDefault output directory for generated images. Defaults to ~/Pictures/chatgpt-images.
CHATGPT_IMAGE_TIMEOUT_MSNoTimeout for each call in milliseconds. Defaults to 600000.

Instructions

Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.

This server publishes no instructions, or was last inspected before Glama recorded them.

Capabilities

Features and capabilities supported by this server

Protocol revision2025-11-25

CapabilityDetails
tools
{
  "listChanged": true
}

Tools

Functions exposed to the LLM to take actions

NameDescription
generate_imageA

Generate images with ChatGPT's image generator, through the ChatGPT/Codex app installed on this machine and the user's ChatGPT account (no API key). Local images can be passed as style/composition references. Each call takes about 30-90 seconds. Saves the files to disk and returns their paths plus a preview.

edit_imageA

Edit an existing image with ChatGPT's image generator (e.g. change background or colors, remove or add objects), through the ChatGPT/Codex app installed on this machine. The original file is never modified: the result is saved as a new file.

Prompts

Interactive templates invoked by user choice

NameDescription

No prompts

Resources

Contextual data attached and managed by the client

NameDescription

No resources

TDQS

A3.9/5.0

Scored across 2 tools

Disambiguation4/5

generate_image creates new images from prompts, while edit_image modifies existing files, so their core purposes are clearly distinct. The only mild overlap is that generate_image accepts local images as style/composition references, which could sometimes resemble editing, but the descriptions clarify the boundary.

Naming Consistency5/5

Both tools follow the exact same verb_noun convention: generate_image and edit_image. The names are predictable and immediately readable.

Tool Count4/5

Two tools cover the core generation and editing operations for an image-generation MCP server, and both earn their place. The set is slightly thin compared with the typical 3-15 range, but it is reasonable for this narrowly focused wrapper.

Completeness4/5

The surface covers the two primary operations: creating new images and editing existing ones. Minor gaps remain around variations, upscaling, or batch operations, but the described workflows are usable without dead ends.

Maintenance

ActivityNo data
ResponsivenessOutdated data