GPT Image Skill

Generate and edit GPT images from Codex, Claude Code, Google Antigravity, or another compatible local agent through the user's ChatGPT subscription. Direct prompts stay unchanged; when the user delegates several different designs, the agent develops a distinct image-ready prompt for each concept. Real reference files pass into generation, bridge-produced transparent PNGs are checked for alpha/transparency support, results stay in the active project, and ready outputs run with bounded parallelism.

install skill → Sign in with ChatGPT → direct prompt or delegated concept prompts + local image inputs
              → single generate or bounded parallel batch
              → built-in $imagegen → <project>/generated-images/*.png

This repository does not call the OpenAI Images API and does not create a separately billed Images API request. Built-in image generation still consumes included ChatGPT/Codex usage and remains subject to plan and workspace limits. OpenAI currently states that Codex image generation is not available on the Free plan and that supported-plan image generations use included limits 3–5× faster on average than similar non-image turns, depending on quality and size. This skill does not bypass that boundary.

Install by pasting one prompt into an agent

Paste this into Codex, Claude Code, Google Antigravity, or another local coding agent:

Install and verify GPT Image Skill for the current user from:

https://github.com/GENEXIS-AI/gpt-image-skill

For this task, I authorize read-only environment checks; a persistent clone or safe fast-forward update;
user-level installation of missing Git, a supported Node.js 22+ LTS, and Codex CLI;
creation of the gpt-image links for Codex, Claude Code, and Google Antigravity; and starting Sign in with ChatGPT device authorization.

Read AGENT_INSTALL.md at the repository root and follow it as the one-time installation contract.
Do not use the Images API, OPENAI_API_KEY, or API-key login. Do not generate a live image yet.
Pause only if administrator privileges are required, an unrelated existing path would be changed,
local changes would be discarded, or existing Codex authentication would need to be replaced.
Otherwise, install the required components, run bootstrap --target all --yes --json,
and continue until bootstrap's consolidated readiness report shows best_practice_pass=true.
Do not add a separate doctor, plan, inspect, or no-image generation check when bootstrap passes.
Finally, report the persistent clone path, all three installed skill paths, and ChatGPT-auth evidence.
Then give me the brief getting_started guide in my language: common aspect ratios,
quality phrases, one creation example, one reference or revision example, and one transparent-background example.
Do not use unexplained jargon such as "dry-run"; call it a setup check that does not create an image.

This prompt authorizes ordinary user-level setup without authorizing administrator elevation, destructive changes, replacement of existing authentication, a live generation, or a GitHub Star. The full boundary is in AGENT_INSTALL.md.

After setup, invoke $gpt-image in Codex, /gpt-image in Claude Code, or mention gpt-image in an Antigravity request. Antigravity discovers the global skill automatically from its official skill directory. The host loads the concise skill only for image tasks; it does not need to reread this README on every request.

What the agent shows after installation

The installing agent should end with a small guide like this, translated into the user's language:

GPT Image Skill is ready. No image was generated during setup.

Common aspect-ratio requests: 1:1, 16:9, 9:16, 4:3, 3:4
Quality phrases: draft, high quality, high detail / final quality

Try:
$gpt-image Create a cozy reading room at sunset, 16:9, high quality.
/gpt-image Use @references/character.png as the character reference and place it in a rainy city, 9:16, high quality.
Use the gpt-image skill to create a flat blue robot app icon with a transparent background, 1:1, high quality.

CLI model policy: no model ID is pinned; Codex selects a current account-available model at Low reasoning.
Plan note: setup verifies sign-in, not image entitlement. Current Codex pricing excludes image generation from Free.

These are common natural-language requests, not a fixed API size list. Other framing or dimension requests can be written normally, and exact pixel dimensions may vary with built-in image generation. The guide appears once after installation rather than after every image.

Design principles

1. The prompt is authoritative—and so is delegated creative intent

For one direct image or edit, the skill forwards the user's image request unchanged. It does not “improve” a short prompt with unsolicited details. A request such as “make five different poster designs” is different: it explicitly asks the agent to develop five creative concepts. The agent preserves the shared subject, references, brand, text, ratio, and other constraints, then sends one complete, meaningfully different image prompt per output. Job numbers remain in IDs and filenames, never in image prompts.

2. References are files, not descriptions

The bridge requires a readable local PNG, JPEG, or WebP path for each reference. It passes those files into the actual $imagegen call. It never replaces an unresolved reference with a text description and continues anyway.

In Claude Code or Antigravity:

• Prefer @path/to/image.png or an explicit filesystem path.
• A pasted or dragged image visible in the conversation is not automatically inherited by a nested codex exec process.
• If the host exposes an exact readable temporary attachment path, the agent copies that exact file into <project>/generated-images/inputs/ and uses the copy.
• If the host exposes no path, save the image inside the project and provide that path before generation.
• The skill does not guess from ~/.claude/image-cache; choosing “the newest image” could select the wrong or private file.

3. Revisions always edit the latest result

Each bridge invocation is ephemeral. For “change the result you just made,” the agent must use the previous generated output as the next --edit-target and reattach every still-needed reference. Reusing the original source is a different operation and loses the prior edit.

4. Normal generation stays light

The default path is:

quick ChatGPT-auth check → current account-available Codex model at Low reasoning
                         → built-in image generation → minimal PNG sanity check
                         → PATH + inline Markdown

Planning, the setup check that does not create an image (--dry-run), capabilities --json, inspect --input, and detailed JSON remain available for troubleshooting. They are not required before a normal image request.

5. The model stays current; reasoning stays light

Codex model names and availability change over time. The default CLI bridge therefore does not pin Luna, Sol, Terra, or any other model ID and does not embed a model catalog. Codex selects a current model available to the signed-in account, while the runner requests Low reasoning—the CLI name for Light in the ChatGPT app. The selected Codex model receives a finalized, tightly scoped instruction and calls $imagegen.

The image renderer is not pinned either. OpenAI currently documents Codex's built-in renderer as gpt-image-2, but this repository invokes the built-in image-generation capability by name so OpenAI can update the underlying renderer without a skill release. The outer Codex model can affect tool routing and instruction following, but it does not replace the built-in renderer or directly set its visual rendering quality.

The default auto policy is intentionally small:

1. Let Codex select a current model available to the signed-in account.
2. Request Low reasoning for the deterministic bridge task.
3. Run no model-discovery preflight, fallback turn, or automatic image-generation retry.

--orchestrator-model account-default removes even the Low override and leaves both model and reasoning to Codex. A current model ID can still be pinned as an advanced, explicit user choice; the runner accepts the ID without maintaining its own allowlist.

This policy lowers bridge overhead on supported plans, but it cannot make image generation available on ChatGPT Free. Codex itself is included in Free, while the official pricing page separately says image generation is unavailable there. See Codex models, Codex image generation, and Codex pricing.

6. Parallelism is explicit and bounded

generate remains the one-image happy path. For two or more outputs, the skill automatically batches every job whose inputs already exist. Different design concepts run independently; same-design variants may run together while reading one shared edit target or design reference. Only an output-to-input dependency creates another stage. The batch checks ChatGPT auth once and uses default concurrency 2, maximum 4, with no Doctor, planning, inspection, model-discovery preflight, fallback turn, or image-generation retry per job.

Features

• Uses Codex's built-in $imagegen under Sign in with ChatGPT
• Uses a host-native OpenAI/Codex image_gen tool directly only when it is backed by included ChatGPT/Codex usage
• Blocks `