AI turns documents or topics into real, native PowerPoint decks—with native shapes, transitions and animations, data-backed charts and tables on demand, audio narration from speaker notes, and support for your own .pptx templates. · by Hugo He
fix(openai): avoid sdk transport for image generation
hugohe3aa1215f10f9f
diff --git a/skills/ppt-master/requirements.txt b/skills/ppt-master/requirements.txtindex 1259c053..8a6a3ab0 100644--- a/skills/ppt-master/requirements.txt+++ b/skills/ppt-master/requirements.txt@@ -100,8 +100,8 @@ curl_cffi>=0.7.0# ─────────────────────────────────────────────────────────────# Gemini backend / Gemini 后端 (IMAGE_BACKEND=gemini)google-genai>=1.0.0-# OpenAI-compatible backend / OpenAI 兼容后端 (IMAGE_BACKEND=openai)-openai>=1.0.0+# OpenAI-compatible backend uses requests from the web tooling section above.+# OpenAI 兼容后端复用上方网页工具所需的 requests。# ─────────────────────────────────────────────────────────────# SVG Editor / SVG 编辑器 (skills/ppt-master/scripts/svg_editor/server.py)diff --git a/skills/ppt-master/scripts/docs/image.md b/skills/ppt-master/scripts/docs/image.mdindex ecd3d3c1..6de16dde 100644--- a/skills/ppt-master/scripts/docs/image.md+++ b/skills/ppt-master/scripts/docs/image.md@@ -101,6 +101,8 @@ Current process environment wins over `.env`.OpenAI backend notes:- `gpt-image-2` is the default OpenAI model.+- Requests are sent with plain `requests.post()` to improve compatibility with+ OpenAI-compatible proxies that block the OpenAI SDK's `httpx` transport.- For `gpt-image-2`, `image_size=512px` means a low-quality draft preset, not a literal 512px edge. The model requires both edges to be multiples of 16px, a long:short ratio no greater than 3:1, and total pixels between 655,360 and 8,294,400.- `OPENAI_BACKGROUND=transparent` is not supported by `gpt-image-2`; use `auto` or `opaque`.- If `OPENAI_OUTPUT_FORMAT=jpeg` or `webp`, generated files use `.jpg` or `.webp` extensions instead of `.png`.diff --git a/skills/ppt-master/scripts/docs/troubleshooting.md b/skills/ppt-master/scripts/docs/troubleshooting.mdindex f53ef502..9b177146 100644--- a/skills/ppt-master/scripts/docs/troubleshooting.md+++ b/skills/ppt-master/scripts/docs/troubleshooting.md@@ -77,4 +77,4 @@ Important optional packages:- `Pillow` for image utilities- `numpy` for watermark removal- `PyMuPDF` for PDF conversion-- `google-genai` / `openai` for image generation backends+- `google-genai` for Gemini image generationdiff --git a/skills/ppt-master/scripts/image_backends/backend_openai.py b/skills/ppt-master/scripts/image_backends/backend_openai.pyindex 6c835645..4ee19288 100644--- a/skills/ppt-master/scripts/image_backends/backend_openai.py+++ b/skills/ppt-master/scripts/image_backends/backend_openai.py@@ -15,7 +15,7 @@ Configuration keys:OPENAI_MODERATION (optional) auto or low for GPT image modelsDependencies:- pip install openai Pillow+ pip install requests Pillow"""import sys@@ -31,10 +31,11 @@ import timeimport threadingfrom collections.abc import Mapping-from openai import OpenAI+import requestsfrom image_backends.backend_common import (MAX_RETRIES,download_image,+ http_error,is_rate_limit_error,normalize_image_size,resolve_output_path,@@ -137,6 +138,7 @@ GPT_IMAGE_OUTPUT_EXTENSIONS = {}GPT_IMAGE_BACKGROUNDS = {"auto", "opaque", "transparent"}GPT_IMAGE_MODERATION_VALUES = {"auto", "low"}+DEFAULT_BASE_URL = "https://api.openai.com/v1"def _field(value, name: str):@@ -264,6 +266,29 @@ def _gpt_image_options(model: str) -> tuple[dict, str]:return options, output_ext+def _image_generations_url(base_url: str | None) -> str:+ return f"{(base_url or DEFAULT_BASE_URL).rstrip('/')}/images/generations"+++def _post_image_generation(api_key: str, base_url: str | None, request: dict) -> dict:+ headers = {+ "Authorization": f"Bearer {api_key}",+ "Content-Type": "application/json",+ }+ response = requests.post(+ _image_generations_url(base_url),+ headers=headers,+ json=request,+ timeout=300,+ )+ if not response.ok:+ raise http_error(response, "OpenAI image generation")+ try:+ return response.json()+ except ValueError as exc:+ raise RuntimeError("OpenAI image generation returned invalid JSON.") from exc++# ╔══════════════════════════════════════════════════════════════════╗# ║ Image Generation ║# ╚══════════════════════════════════════════════════════════════════╝@@ -283,8 +308,6 @@ def _generate_image(api_key: str, prompt: str,Raises:RuntimeError: When generation fails"""- client = OpenAI(api_key=api_key, base_url=base_url)-# Map parameterssize = _select_size(model, aspect_ratio, image_size)quality = IMAGE_SIZE_TO_QUALITY.get(image_size, "auto")@@ -335,7 +358,7 @@ def _generate_image(api_key: str, prompt: str,hb_thread.start()try:- resp = client.images.generate(**request)+ resp = _post_image_generation(api_key, base_url, request)finally:heartbeat_stop.set()hb_thread.join(timeout=1)diff --git a/skills/ppt-master/scripts/image_gen.py b/skills/ppt-master/scripts/image_gen.pyindex 01b81f78..2e9ab3de 100644--- a/skills/ppt-master/scripts/image_gen.py+++ b/skills/ppt-master/scripts/image_gen.py@@ -6,7 +6,7 @@ Dispatches to the appropriate backend based on explicit provider configuration.Backend selection (`IMAGE_BACKEND` in `.env` or the current process environment):IMAGE_BACKEND=gemini -> Gemini backend (google-genai SDK)- IMAGE_BACKEND=openai -> OpenAI-compatible backend (openai SDK)+ IMAGE_BACKEND=openai -> OpenAI-compatible backend (raw HTTP via requests)IMAGE_BACKEND=minimax -> MiniMax image backendIMAGE_BACKEND=stability -> Stability AI backendIMAGE_BACKEND=bfl -> Black Forest Labs FLUX backend