[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-80739":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":9,"language":10,"languages":9,"totalLinesOfCode":9,"stars":11,"forks":12,"watchers":13,"openIssues":14,"contributorsCount":14,"subscribersCount":14,"size":14,"stars1d":14,"stars7d":12,"stars30d":15,"stars90d":14,"forks30d":14,"starsTrendScore":14,"compositeScore":16,"rankGlobal":9,"rankLanguage":9,"license":17,"archived":18,"fork":18,"defaultBranch":19,"hasWiki":18,"hasPages":18,"topics":20,"createdAt":9,"pushedAt":9,"updatedAt":21,"readmeContent":22,"aiSummary":23,"trendingCount":14,"starSnapshotCount":14,"syncStatus":24,"lastSyncTime":25,"discoverSource":26},80739,"EditableImage2PPTSkill","soulmujoco\u002FEditableImage2PPTSkill","soulmujoco","Codex skill for converting PPT slide images into editable PowerPoint decks",null,"Python",47,3,41,0,6,1.81,"MIT License",false,"main",[],"2026-06-12 02:04:06","# EditableImage2PPTSkill\n\nCodex skill for reconstructing slide screenshots or exported slide images into editable PowerPoint decks.\n\nThis skill is designed for cases where the source material is an image of a slide rather than a native `.pptx` file. It helps Codex extract reusable visual assets, rebuild text and geometry as editable PowerPoint objects, and package the result as a clean `.pptx`.\n\n## What It Does\n\n- Converts one or more slide images into editable PowerPoint slides.\n- Keeps ordinary text as editable PowerPoint text boxes.\n- Rebuilds simple frames, cards, circles, separators, and rules as native shapes.\n- Extracts logos, icons, photos, decorative bands, and complex visual elements as independent PNG assets.\n- Supports batch reconstruction by creating one layout JSON file per page and combining them into a final deck.\n- Includes QA helpers for inspecting generated PPTX files.\n\n## Repository Layout\n\n```text\n.\n├── SKILL.md\n├── agents\u002F\n│   └── openai.yaml\n├── references\u002F\n│   ├── asset-manifest.md\n│   ├── layout-json.md\n│   └── reconstruction-sop.md\n└── scripts\u002F\n    ├── asset_cropper.py\n    ├── build_pptx_from_layout.py\n    ├── combine_layouts.py\n    └── inspect_pptx.py\n```\n\n## Installation\n\nCopy this repository into your Codex skills directory:\n\n```powershell\ngit clone https:\u002F\u002Fgithub.com\u002Fsoulmujoco\u002FEditableImage2PPTSkill.git \"$env:USERPROFILE\\.codex\\skills\\EditableImage2PPTSkill\"\n```\n\nThe skill will be available to Codex after it reloads skills.\n\n## Requirements\n\nUse an active Python environment with:\n\n- `python-pptx`\n- `Pillow`\n- `numpy`\n\nIf needed:\n\n```bash\npython -m pip install python-pptx pillow numpy\n```\n\n## Typical Workflow\n\n1. Place source slide images in a working folder.\n2. Create an asset crop manifest for each page.\n3. Run `scripts\u002Fasset_cropper.py` to extract PNG assets and generate a contact sheet.\n4. Create layout JSON files using source-image pixel coordinates.\n5. Run `scripts\u002Fbuild_pptx_from_layout.py` to generate editable slides.\n6. Use `scripts\u002Fcombine_layouts.py` for multi-page decks.\n7. Run `scripts\u002Finspect_pptx.py` to check the final PPTX package.\n\n## Example Commands\n\nExtract visual assets:\n\n```bash\npython scripts\u002Fasset_cropper.py \\\n  --manifest page_002.assets.json \\\n  --out-dir assets\u002Fpage_002 \\\n  --contact-sheet\n```\n\nBuild a PPTX from a layout:\n\n```bash\npython scripts\u002Fbuild_pptx_from_layout.py \\\n  --layout layouts\u002Fpage_002.layout.json \\\n  --assets-root . \\\n  --out output\u002Fpage_002_editable.pptx\n```\n\nCombine layouts:\n\n```bash\npython scripts\u002Fcombine_layouts.py \\\n  --layouts layouts \\\n  --out layouts\u002Fcombined.layout.json\n```\n\nInspect a generated deck:\n\n```bash\npython scripts\u002Finspect_pptx.py \\\n  --pptx output\u002Fdeck_editable.pptx \\\n  --report scratch\u002Fquality_report.json\n```\n\n## Reconstruction Standard\n\nA reconstruction is considered complete when:\n\n- Meaningful visual assets are either extracted as PNGs or rebuilt as native PowerPoint shapes.\n- Ordinary text remains editable.\n- Extracted assets are separate picture objects.\n- The final PPTX passes package inspection.\n- The generated preview has been visually compared against the source images, or any remaining fidelity limits are explicitly documented.\n\n## Notes\n\nThis skill is for image-to-PPT reconstruction. If the original native `.pptx` file is available, edit that deck directly instead of reconstructing it from screenshots.\n\n## License\n\nMIT License. See [LICENSE](LICENSE) for details.\n","EditableImage2PPTSkill 是一个用于将 PPT 幻灯片图像转换为可编辑的 PowerPoint 文档的 Codex 技能。其核心功能包括将图片中的文本保持为可编辑的文字框，重建简单图形如边框、卡片等为原生形状，并提取标志、图标等复杂视觉元素作为独立的 PNG 资源。该技能支持批量处理，通过创建每页一个布局 JSON 文件并合并生成最终文档。适用于仅有幻灯片截图或导出图片而无原始 `.pptx` 文件的情况，帮助用户恢复和编辑内容。项目基于 Python 开发，依赖于 `python-pptx`、`Pillow` 和 `numpy` 库，适合需要从静态图像中恢复 PowerPoint 内容的场景使用。",2,"2026-06-11 04:01:51","CREATED_QUERY"]