Skip to content
← Directory

skill · Global

codex-media-gen

A Claude skill that lets Claude generate real AI raster images by shelling out to the user's locally-installed Codex CLI (gpt-image-2) through Desktop Commander MCP.

Feature or claim this listing

Feature this listing — founding Featured slots are 50% off for 6 months

Represent codex-media-gen? Send a claim for review so we can discuss listing updates. Claims do not automatically grant ownership or verification.

Explore paid sponsorship →

What's in it

Data offered docs-verified
AI raster image generation (PNG) via gpt-image-2, run through local Codex CLI
Formats docs-verified
PNG
Access docs-verified
MIT licence; requires a ChatGPT Plus or Pro subscription (for Codex gpt-image-2), a locally installed Codex CLI, and the Desktop Commander MCP connected to Claude

Useful for

Use cases llm-inferred
Illustrative cover image for a flood risk assessment or drainage strategy report · Concept illustration of a SuDS or natural flood management scheme for a client presentation · Hero or banner image for a consultancy project page or proposal
Audience llm-inferred
consultancy communications and marketing staff · report authors and presentation designers · non-technical staff needing illustrative (non-data) graphics

Limitations

Constraints llm-inferred
Produces illustrative raster images only; provides no water, civils or environmental data · Requires Codex CLI v0.130.0+, ChatGPT Plus or Pro auth, and image_generation = true in ~/.codex/config.toml · Requires Desktop Commander MCP connected to Claude · Setup described for macOS paths; installation steps depend on the Claude client · Generation takes about 45-75 seconds per image · Batch generation, retry loops, video, image editing, and transparent backgrounds are out of scope · Skill loading is in active development, so behaviour may change across clients
Licence
MIT
Authentication
Unknown — review source docs
GitHub stars
1
Last commit
16/05/2026
Last successful page check
14/10/2026
Page health
healthy
Package
npm:@openai/codex
Verification
Not independently verified

Auto-extracted, unverified fields: tools, install, auth. Check upstream documentation.

Provenance & evidence

Tools/install/auth auto-extracted from README; quality score heuristic (stars, recency, tool count, licence).

Metadata observed 16/10/2026 · research sweep: github-search