gpt-image-2

Name: gpt-image-2
Author: GargantuaX

4stars0forks

How to install gpt-image-2?

gargantuax/openskills/gpt-image-2

$npx -y skills add GargantuaX/openskills --skill gpt-image-2

Installs into the current project.

Summary

Full OpenAI-compatible GPT Image 2 coverage across images/generations, images/edits, and responses with the image_generation tool. Use when the one-shot image helper is not enough - text-to-image, mask edits, multi-image batches, streaming, partial_images, and mixed text+image Re

Files · 1

View on GitHub

SKILL.md

1	# GPT Image 2
2
3	A single Python entrypoint that covers every GPT Image 2 route, with strict pre-flight validation of the model's size, aspect, and feature constraints.
4
5	## Workflow
6
7	1. Open [references/config.md](./references/config.md) to pick environment variables and defaults.
8	2. Open [references/api-surface.md](./references/api-surface.md) to choose between `generations`, `edits`, and `responses`.
9	3. Prefer `OPENAI_BASE_URL=https://api.openai.com/v1` unless the user asks for a different OpenAI-compatible endpoint.
10	4. Use `gpt-image-2` for `generations` and `edits`; use a text-capable Responses model such as `gpt-5.4` for `responses`.
11	5. Run `scripts/gpt_image.py` with one of the three subcommands.
12	6. Add `--dry-run` first when the payload shape is the main risk.
13	7. Add `--save-response <path>` when the raw JSON body or SSE event stream needs to be kept for debugging.
14
15	## Commands
16
17	Text-to-image through the public Images API:
18
19	```powershell
20	python .\skills\gpt-image-2\scripts\gpt_image.py generations `
21	--prompt "A bold product hero image for a developer tool homepage" `
22	--output .\out\hero.png `
23	--size 1536x1024 `
24	--quality high `
25	--format png
26	```
27
28	Multi-image batch with a filename pattern:
29
30	```powershell
31	python .\skills\gpt-image-2\scripts\gpt_image.py generations `
32	--prompt "A cinematic city skyline at night" `
33	--output .\out\skyline-{index}.webp `
34	--n 3 `
35	--format webp `
36	--compression 90
37	```
38
39	Image edits with two inputs plus a mask:
40
41	```powershell
42	python .\skills\gpt-image-2\scripts\gpt_image.py edits `
43	--prompt "Blend the two references into one clean marketing illustration" `
44	--image .\refs\subject.png `
45	--image .\refs\background.png `
46	--mask .\refs\mask.png `
47	--output .\out\edit-{index}.png `
48	--image-field-style brackets `
49	--n 2
50	```
51
52	Responses API with streaming and partial previews:
53
54	```powershell
55	python .\skills\gpt-image-2\scripts\gpt_image.py responses `
56	--input-text "Generate a poster for an AI developer summit" `
57	--model gpt-5.4 `
58	--output .\out\poster-{index}.png `
59	--stream `
60	--partial-images 2 `
61	--save-response .\out\poster-events.json
62	```
63
64	Responses API edit with a local image plus a mask:
65
66	```powershell
67	python .\skills\gpt-image-2\scripts\gpt_image.py responses `
68	--input-text "Turn this product shot into a clean studio ad" `
69	--model gpt-5.4 `
70	--input-image .\refs\product.png `
71	--mask .\refs\mask.png `
72	--output .\out\studio.png `
73	--action edit
74	```
75
76	Inspect the built request without sending it:
77
78	```powershell
79	python .\skills\gpt-image-2\scripts\gpt_image.py generations `
80	--prompt "A minimal cover image" `
81	--output .\out\cover.png `
82	--dry-run
83	```
84
85	## Rules
86
87	- Use `generations` for public text-to-image calls.
88	- Use `edits` for multipart image edits and mask uploads.
89	- Use `responses` for advanced flows: streaming, mixed text + image input, `previous_response_id`, `tool_choice`, `action`, and optional `tool_model`.
90	- Process environment variables override `.env`; CLI flags override both.
91	- Never print secrets.
92	- `--output` takes either a single path or a pattern such as `image-{index}.png` for multi-image or streaming flows.
93	- `responses` uses a top-level Responses model separate from the image model; default it to `gpt-5.4` unless you need another text-capable model.
94	- `quality` on Responses tool flows is passed through, but final behavior still depends on the hosted image tool.
95	- On OpenAI GPT image models, omit `response_format`; image data already comes back as base64.
96	- Fail fast on unsupported `gpt-image-2` combinations: transparent background, invalid size, `partial_images` outside `0..3`, or `stream=true` with `n>1` on public Images routes.
97
98	## Resources
99
100	- Script: [scripts/gpt_image.py](./scripts/gpt_image.py)
101	- Config reference: [references/config.md](./references/config.md)
102	- API surface reference: [references/api-surface.md](./references/api-surface.md)

People who liked this also viewed

6 picks

Type

gpt-image-2

4stars0forks

View on GitHub

How to install gpt-image-2?

gargantuax/openskills/gpt-image-2

$npx -y skills add GargantuaX/openskills --skill gpt-image-2

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# GPT Image 2
2
3	A single Python entrypoint that covers every GPT Image 2 route, with strict pre-flight validation of the model's size, aspect, and feature constraints.
4
5	## Workflow
6
7	1. Open [references/config.md](./references/config.md) to pick environment variables and defaults.
8	2. Open [references/api-surface.md](./references/api-surface.md) to choose between `generations`, `edits`, and `responses`.
9	3. Prefer `OPENAI_BASE_URL=https://api.openai.com/v1` unless the user asks for a different OpenAI-compatible endpoint.
10	4. Use `gpt-image-2` for `generations` and `edits`; use a text-capable Responses model such as `gpt-5.4` for `responses`.
11	5. Run `scripts/gpt_image.py` with one of the three subcommands.
12	6. Add `--dry-run` first when the payload shape is the main risk.
13	7. Add `--save-response <path>` when the raw JSON body or SSE event stream needs to be kept for debugging.
14
15	## Commands
16
17	Text-to-image through the public Images API:
18
19	```powershell
20	python .\skills\gpt-image-2\scripts\gpt_image.py generations `
21	--prompt "A bold product hero image for a developer tool homepage" `
22	--output .\out\hero.png `
23	--size 1536x1024 `
24	--quality high `
25	--format png
26	```
27
28	Multi-image batch with a filename pattern:
29
30	```powershell
31	python .\skills\gpt-image-2\scripts\gpt_image.py generations `
32	--prompt "A cinematic city skyline at night" `
33	--output .\out\skyline-{index}.webp `
34	--n 3 `
35	--format webp `
36	--compression 90
37	```
38
39	Image edits with two inputs plus a mask:
40
41	```powershell
42	python .\skills\gpt-image-2\scripts\gpt_image.py edits `
43	--prompt "Blend the two references into one clean marketing illustration" `
44	--image .\refs\subject.png `
45	--image .\refs\background.png `
46	--mask .\refs\mask.png `
47	--output .\out\edit-{index}.png `
48	--image-field-style brackets `
49	--n 2
50	```
51
52	Responses API with streaming and partial previews:
53
54	```powershell
55	python .\skills\gpt-image-2\scripts\gpt_image.py responses `
56	--input-text "Generate a poster for an AI developer summit" `
57	--model gpt-5.4 `
58	--output .\out\poster-{index}.png `
59	--stream `
60	--partial-images 2 `
61	--save-response .\out\poster-events.json
62	```
63
64	Responses API edit with a local image plus a mask:
65
66	```powershell
67	python .\skills\gpt-image-2\scripts\gpt_image.py responses `
68	--input-text "Turn this product shot into a clean studio ad" `
69	--model gpt-5.4 `
70	--input-image .\refs\product.png `
71	--mask .\refs\mask.png `
72	--output .\out\studio.png `
73	--action edit
74	```
75
76	Inspect the built request without sending it:
77
78	```powershell
79	python .\skills\gpt-image-2\scripts\gpt_image.py generations `
80	--prompt "A minimal cover image" `
81	--output .\out\cover.png `
82	--dry-run
83	```
84
85	## Rules
86
87	- Use `generations` for public text-to-image calls.
88	- Use `edits` for multipart image edits and mask uploads.
89	- Use `responses` for advanced flows: streaming, mixed text + image input, `previous_response_id`, `tool_choice`, `action`, and optional `tool_model`.
90	- Process environment variables override `.env`; CLI flags override both.
91	- Never print secrets.
92	- `--output` takes either a single path or a pattern such as `image-{index}.png` for multi-image or streaming flows.
93	- `responses` uses a top-level Responses model separate from the image model; default it to `gpt-5.4` unless you need another text-capable model.
94	- `quality` on Responses tool flows is passed through, but final behavior still depends on the hosted image tool.
95	- On OpenAI GPT image models, omit `response_format`; image data already comes back as base64.
96	- Fail fast on unsupported `gpt-image-2` combinations: transparent background, invalid size, `partial_images` outside `0..3`, or `stream=true` with `n>1` on public Images routes.
97
98	## Resources
99
100	- Script: [scripts/gpt_image.py](./scripts/gpt_image.py)
101	- Config reference: [references/config.md](./references/config.md)
102	- API surface reference: [references/api-surface.md](./references/api-surface.md)

People who liked this also viewed

6 picks

Type