feiskyer/codex-settings / skills/gpt-image-skill
gpt-image-skill
VerifiedGenerate or edit images when the user names OpenAI, GPT Image, or a gpt-image model. Use the default image skill when no provider is named.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 2.2 KB |
| agents/openai.yaml | config | 260 B |
| evals/evals.json | config | 926 B |
| gpt_image.pyexecutable | script | 10.2 KB |
| requirements.txt | manifest | 27 B |
| tests/test_gpt_image.py | script | 6.3 KB |
tests/test_gpt_image.py
View on GitHubimport base64
from contextlib import redirect_stderr, redirect_stdout
import io
import importlib.util
from pathlib import Path
from types import SimpleNamespace
import tempfile
import unittest
from unittest import mock
SKILL_DIR = Path(__file__).resolve().parents[1]
MODULE_PATH = SKILL_DIR / "gpt_image.py"
SPEC = importlib.util.spec_from_file_location("gpt_image", MODULE_PATH)
gpt_image = importlib.util.module_from_spec(SPEC)
SPEC.loader.exec_module(gpt_image)
def image_response(content=b"image-bytes"):
return SimpleNamespace(
data=[
SimpleNamespace(
b64_json=base64.b64encode(content).decode("ascii"),
url=None,
revised_prompt=None,
)
]
)
class FakeImages:
def __init__(self, response):
self.response = response
self.edit_params = None
def edit(self, **params):
self.edit_params = params
return self.response
class FakeClient:
def __init__(self, response):
self.images = FakeImages(response)
class ArgumentTests(unittest.TestCase):
def test_legacy_generation_syntax_still_works(self):
args = gpt_image.parse_args(["--prompt", "draw a cat"])
self.assertEqual(args.command, "generate")
self.assertEqual(args.prompt, "draw a cat")
self.assertEqual(args.format, "png")
self.assertTrue(args.output.endswith(".png"))
def test_documented_edit_syntax_requires_prompt_only_once(self):
args = gpt_image.parse_args(
["edit", "--prompt", "add a rainbow", "--input", "photo.png"]
)
self.assertEqual(args.command, "edit")
self.assertEqual(args.prompt, "add a rainbow")
self.assertEqual(args.input, ["photo.png"])
def test_output_format_is_inferred_from_filename(self):
args = gpt_image.parse_args(
["--prompt", "draw a cat", "--output", "cat.webp"]
)
self.assertEqual(args.format, "webp")
def test_mismatched_output_extension_is_rejected(self):
with redirect_stderr(io.StringIO()):
with self.assertRaises(SystemExit):
gpt_image.parse_args(
[
"--prompt",
"draw a cat",
"--format",
"jpeg",
"--output",
"cat.png",
]
)
class EditTests(unittest.TestCase):
def test_edit_passes_format_and_background_and_creates_parent(self):
with tempfile.TemporaryDirectory() as temp_dir:
root = Path(temp_dir)
input_path = root / "input.png"
input_path.write_bytes(b"input")
output_path = root / "nested" / "edited.webp"
args = gpt_image.parse_args(
[
"edit",
"--prompt",
"replace the background",
"--input",
str(input_path),
"--format",
"webp",
"--background",
"transparent",
"--output",
str(output_path),
]
)
client = FakeClient(image_response())
with redirect_stdout(io.StringIO()):
saved_paths = gpt_image.edit_image(client, args)
self.assertEqual(saved_paths, [str(output_path)])
self.assertEqual(output_path.read_bytes(), b"image-bytes")
self.assertEqual(client.images.edit_params["output_format"], "webp")
self.assertEqual(client.images.edit_params["background"], "transparent")
def test_missing_input_fails_before_api_call(self):
args = gpt_image.parse_args(
["edit", "--prompt", "edit", "--input", "/missing/image.png"]
)
client = FakeClient(image_response())
with self.assertRaises(FileNotFoundError):
gpt_image.edit_image(client, args)
self.assertIsNone(client.images.edit_params)
class SaveResultTests(unittest.TestCase):
def test_no_image_content_is_an_error(self):
response = SimpleNamespace(
data=[SimpleNamespace(b64_json=None, url=None, revised_prompt=None)]
)
args = SimpleNamespace(output="unused.png")
with redirect_stderr(io.StringIO()):
with self.assertRaises(RuntimeError):
gpt_image.save_results(response, args)
class DownloadTests(unittest.TestCase):
def test_download_checks_status_and_writes_streamed_content(self):
response = mock.MagicMock()
response.__enter__.return_value = response
response.iter_bytes.return_value = [b"image-", b"bytes"]
with tempfile.TemporaryDirectory() as temp_dir:
output_path = Path(temp_dir) / "nested" / "image.png"
output_path.parent.mkdir(parents=True)
with mock.patch.object(
gpt_image.httpx,
"stream",
return_value=response,
) as stream:
gpt_image.download_image("https://example.test/image", output_path)
response.raise_for_status.assert_called_once_with()
self.assertEqual(output_path.read_bytes(), b"image-bytes")
stream.assert_called_once_with(
"GET",
"https://example.test/image",
follow_redirects=True,
timeout=gpt_image.DOWNLOAD_TIMEOUT_SECONDS,
)
def test_oversized_download_removes_partial_file(self):
response = mock.MagicMock()
response.__enter__.return_value = response
response.iter_bytes.return_value = [b"abc", b"def"]
with tempfile.TemporaryDirectory() as temp_dir:
output_path = Path(temp_dir) / "image.png"
with mock.patch.object(gpt_image, "MAX_DOWNLOAD_BYTES", 5):
with mock.patch.object(
gpt_image.httpx,
"stream",
return_value=response,
):
with self.assertRaises(RuntimeError):
gpt_image.download_image(
"https://example.test/image",
output_path,
)
self.assertFalse(output_path.exists())
if __name__ == "__main__":
unittest.main()
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 471k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |