chatgpt-apps
OfficialScaffolds ChatGPT Apps SDK projects with MCP servers and widgets using current official docs.
Memory
Generate or edit raster images for projects, including photos, mockups, sprites, and transparent cutouts.
Produces new bitmap assets or modifies existing ones using a built-in generation tool that requires no API key. Supports specific workflows like product mockups, UI wireframes, background extraction, and style transfer based on structured prompt specs. Outputs are saved directly to the project workspace or rendered inline for preview, avoiding temporary storage for final deliverables.
Generates or edits images for the current project (for example website assets, game assets, UI mockups, product mockups, wireframes, logo design, photorealistic images, or infographics).
This skill has exactly two top-level modes:
image_gen tool for normal image generation and editing. Does not require OPENAI_API_KEY.scripts/image_gen.py CLI. Use only when the user explicitly asks for the CLI path. Requires OPENAI_API_KEY.Within the explicit CLI fallback only, the CLI exposes three subcommands:
generateeditgenerate-batchRules:
image_gen tool by default for all normal image generation and editing requests.OPENAI_API_KEY. Proceed only if the user explicitly asks for that fallback.scripts/image_gen.py workflow. Do not create one-off SDK runners.scripts/image_gen.py. If something is missing, ask the user before doing anything else.Built-in save-path policy:
$CODEX_HOME/* by default.image_gen tool. If a specific location is needed, generate first and then move or copy the selected output from $CODEX_HOME/generated_images/....$CODEX_HOME/* path.$CODEX_HOME/* path.hero-v2.png or item-icon-edited.png.Shared prompt guidance for both modes lives in references/prompting.md and references/sample-prompts.md.
Fallback-only docs/resources for CLI mode:
references/cli.mdreferences/image-api.mdreferences/codex-network.mdscripts/image_gen.pyThink about two separate questions:
Intent:
Built-in edit semantics:
view_image tool so the image is visible in the conversation context, then proceed with the built-in edit flow.Execution strategy:
image_gen call per requested asset or variant.generate-batch subcommand only when the user explicitly chose CLI mode and needs many prompts/assets.Assume the user wants a new image unless they clearly ask to change an existing one.
generate or edit.generate-batch.view_image first so the image is available in conversation context.image_gen rather than substituting SVG/HTML/CSS placeholders. If the request is for an icon, logo, or UI graphic that should match existing repo-native SVG/vector/code assets, prefer editing those directly instead.image_gen tool by default.input_fidelity, masks, output format, output paths, and network setup.$CODEX_HOME/generated_images/... path.$CODEX_HOME/generated_images/... path.Reformat user prompts into a structured, production-oriented spec. Make the user's goal clearer and more actionable, but do not blindly add detail.
Treat this as prompt-shaping guidance, not a closed schema. Use only the lines that help, and add a short extra labeled line when it materially improves clarity.
Use the user's prompt specificity to decide how much augmentation is appropriate:
Allowed augmentations:
Not allowed augmentations:
Classify each request into one of these buckets and keep the slug consistent across prompts and references.
Generate:
Edit:
Use the following labeled spec as shared prompt scaffolding for both top-level modes:
Use case:
Asset type:
Primary request:
Input images: (optional)
Scene/backdrop:
Subject:
Style/medium:
Composition/framing:
Lighting/mood:
Color palette:
Materials/textures:
Text (verbatim): ""
Constraints:
Avoid:
Notes:
Asset type and Input images are prompt scaffolding, not dedicated CLI flags.Scene/backdrop refers to the visual setting. It is not the same as the fallback CLI background parameter, which controls output transparency behavior.Quality:, Input fidelity:, masks, output format, and output paths belong in the explicit CLI path only. Do not treat them as built-in image_gen tool arguments.Augmentation rules:
change only X; keep Y unchanged).Use case: product-mockup
Asset type: landing page hero
Primary request: a minimal hero image of a ceramic coffee mug
Style/medium: clean product photography
Composition/framing: wide composition with usable negative space for page copy if needed
Lighting/mood: soft studio lighting
Constraints: no logos, no text, no watermark
Use case: precise-object-edit
Asset type: product photo background replacement
Primary request: replace only the background with a warm sunset gradient
Constraints: change only the background; keep the product and its edges unchanged; no text; no watermark
references/cli.md and references/image-api.md for quality, input_fidelity, masks, output format, and output-path guidance.More principles shared by both modes: references/prompting.md.
Copy/paste specs shared by both modes: references/sample-prompts.md.
Asset-type templates (website assets, game assets, wireframes, logo) are consolidated in references/sample-prompts.md.
These conventions apply only to the explicit CLI fallback. They do not describe built-in image_gen output behavior.
tmp/imagegen/ for intermediate files (for example JSONL batches); delete them when done.output/imagegen/.--out or --out-dir to control output paths; keep filenames stable and descriptive.Prefer uv for dependency management in this repo.
Required Python package:
uv pip install openai
Optional for downscaling only:
uv pip install pillow
Portability note:
uv pip install ... remains the preferred path.OPENAI_API_KEY must be set for live API calls.OPENAI_API_KEY when using the built-in image_gen tool.If the key is missing, give the user these steps:
OPENAI_API_KEY as an environment variable in their system.If installation is not possible in this environment, tell the user which dependency is missing and how to install it into their active environment.
references/cli.mdreferences/image-api.mdreferences/codex-network.mdreferences/prompting.md: shared prompting principles for both modes.references/sample-prompts.md: shared copy/paste prompt recipes for both modes.references/cli.md: fallback-only CLI usage via scripts/image_gen.py.references/image-api.md: fallback-only API/CLI parameter reference.references/codex-network.md: fallback-only network/sandbox troubleshooting for CLI mode.scripts/image_gen.py: fallback-only CLI implementation. Do not load or use it unless the user explicitly chooses CLI mode.Scaffolds ChatGPT Apps SDK projects with MCP servers and widgets using current official docs.
Builds or updates full Figma screens using published design system components, variables, and styles instead of hardcoded primitives.
Orchestrates multi-phase Figma design system builds from codebases with mandatory validation checkpoints.
Enforces mandatory Figma Plugin API rules and incremental workflows to prevent script failures during file modifications.
Generates Codex-compatible animated pet spritesheets and packaging from concepts, brand cues, or reference images.
Retrieves current OpenAI developer documentation with citations for API, model selection, and Codex guidance.