Markitdown
ImCa0/just-laws
Convert files and office documents to Markdown. An agent skill from ImCa0/just-laws.
将位图图形重建为高保真可编辑 SVG 与原生 PowerPoint,支持仅文字可编辑、文字/结构/公式可编辑、源图素材保真混合、选择性或完整前景拆分,以及复杂连续背景的 AI 清版底重建。当用户希望把截图、论文插图、流程图、架构图、信息图、UI、图表/地图、海报或封面转成可编辑图形,或提到“图片转可编辑 SVG/PPT”“仅文字可编辑”“图表重建”“FigEdit”时使用。
$ npx skills add giszzt/figedit --skill figedit-v2 -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install giszzt/figedit figedit-v2 --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
Claude Code skills documentation · loads skills from .claude/skills/
Install the "figedit-v2" agent skill from https://github.com/giszzt/figedit/tree/main into .claude/skills/figedit-v2/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "figedit-v2", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add giszzt/figedit --skill figedit-v2 -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install giszzt/figedit figedit-v2 --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "figedit-v2" agent skill from https://github.com/giszzt/figedit/tree/main into .agents/skills/figedit-v2/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "figedit-v2", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add giszzt/figedit --skill figedit-v2 -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install giszzt/figedit figedit-v2 --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "figedit-v2" agent skill from https://github.com/giszzt/figedit/tree/main into .cursor/skills/figedit-v2/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "figedit-v2", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add giszzt/figedit --skill figedit-v2 -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install giszzt/figedit figedit-v2 --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "figedit-v2" agent skill from https://github.com/giszzt/figedit/tree/main into .gemini/skills/figedit-v2/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "figedit-v2", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install giszzt/figedit figedit-v2Installs for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add giszzt/figedit --skill figedit-v2 -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "figedit-v2" agent skill from https://github.com/giszzt/figedit/tree/main into .github/skills/figedit-v2/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "figedit-v2", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add giszzt/figedit --skill figedit-v2 -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install giszzt/figedit figedit-v2 --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "figedit-v2" agent skill from https://github.com/giszzt/figedit/tree/main into .opencode/skills/figedit-v2/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "figedit-v2", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
figedit-v2将位图图形重建为高保真可编辑 SVG 与原生 PowerPoint,支持仅文字可编辑、文字/结构/公式可编辑、源图素材保真混合、选择性或完整前景拆分,以及复杂连续背景的 AI 清版底重建。当用户希望把截图、论文插图、流程图、架构图、信息图、UI、图表/地图、海报或封面转成可编辑图形,或提到“图片转可编辑 SVG/PPT”“仅文字可编辑”“图表重建”“FigEdit”时使用。
Figedit V2 is an agent skill from giszzt/figedit. 将位图图形重建为高保真可编辑 SVG 与原生 PowerPoint,支持仅文字可编辑、文字/结构/公式可编辑、源图素材保真混合、选择性或完整前景拆分,以及复杂连续背景的 AI 清版底重建。当用户希望把截图、论文插图、流程图、架构图、信息图、UI、图表/地图、海报或封面转成可编辑图形,或提到“图片转可编辑 SVG/PPT”“仅文字可编辑”“图表重建”“FigEdit”时使用。
Its SKILL.md is about 2.5k tokens, which your agent loads only when the skill is triggered. The skill folder holds 393 other files, including scripts, reference files and assets (for example `CHANGELOG.md`, `README.en.md` and `README.md`).
It sits in Documents & Office, covering PowerPoint presentations. It works with Microsoft PowerPoint. The repository describes itself as: Turn raster figures into editable SVG and native PowerPoint files. The licence is MIT.
4 steps, taken from the first numbered list in SKILL.md.
Read from SKILL.md and the folder at commit 4450b6b. It shows what the files ask for, not the result of running them.
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
Ships 1 file in scripts/, which the agent can run.
Shell commands in SKILL.md call:
pythonFrom the folder's file list and the shell code blocks in SKILL.md.
No URLs in SKILL.md.
From URLs in SKILL.md, links to its own repository left out.
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Figedit V2 loads about 2.5k tokens when it runs, and up to ~33k if it reads all its reference files. Until then it costs about 50 tokens; SKILL.md has 364 words of instructions outside code blocks.
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
The automated check found no risky patterns in SKILL.md.
Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); the scripts in this folder are not scanned.
The full file from giszzt/figedit at commit 4450b6b, republished under its MIT licence (© giszzt). 364 words, ~2,520 tokens.
.claude/skills/figedit-v2/SKILL.md (or your agent's skills folder). This skill also uses 391 other files; get the full folder from GitHub.把不可编辑的位图图形重建为高保真、可维护、可继续编辑的 SVG 与原生 PPTX 包。
**主流程:勘察 → 备料 → 组装 → 验收。**先判断,后执行。判断阶段不碰坐标也不跑脚本,只回答“每一部分从哪来”;执行阶段才按已定路线调工具。
输出画面上的每一块像素只可能有四个来源:
croppreserve-raster路由就是给每一部分指派来源,每部分有且只有一个。指派完成即可开工;指派不全就是漏了东西。
只有“用三五笔基础图元能画得像”的图形才允许 redraw:朴素箭头、方框、分隔线、圆点、加减号、对勾、简单节点圆。有定制轮廓、渐变、阴影、多色细节、品牌或角色身份的对象一律按素材处理(切、再生或压平),哪怕它看起来简单。拿不准时按素材处理,不重画。重画产物必须与源图形状可对照,画不像就是路由错误,不是绘画技术问题。
所有命令在用户项目目录运行,每张图建任务目录(figure-task/work 证据、figure-task/out 交付),绝不写回 skill 目录。Windows 路径用正斜杠,非 ASCII 路径先经文件系统验证。
看一次足够清晰的源图,产出一份重建路径概要。这是本阶段唯一的交付物,纯文字,零坐标,二十行以内,写给人看。槽位固定,缺项写“无”:
图型与分区 每个区域一行:这是什么、底子从哪来
需要生成的 清版底张数 + 色键版张数 = 计费调用总数,或“无”
不可重画的对象 可直接裁的一堆、需再生的一堆,按名字和位置点出来,或“无”
公式 有无、大致数量 → 验证档位
工作量重心 这张图的时间主要花在哪
需要用户定 具体问题,或“无”对象用名字和空间关系指认(“标题右边那个地球仪,骑在面板边框上”),不写 bbox。坐标属于备料阶段。
概要写完路线即锁定,后续不翻案。写不出“工作量重心”说明没看懂图;“不可重画的对象”分不出两堆说明污染扫描没做。
| 看到的 | 底子从哪来 | 里面的对象从哪来 |
|---|---|---|
| 白底或平色底的流程图、架构图、论文插图 | SVG 画 | 通用图元画;专有图标按窗口干净与否切或再生 |
| 多面板拼接的复合图 | 每个面板分别查 | 同上,按面板成批扫 |
| 截图、图表主体、缩略图网格,内部不需要改 | 整块保留 | 不拆 |
| 地图、照片、插画上压着要改的标注 | 区域 AI 清版 | 要能单独动的再生;不用动的压平进底板 |
| 海报、封面、场景图,文字压在连续画面上 | AI 清版 | 同上 |
| 公式密集的论文图 | 按上面几行查底子 | 公式一律 math,验证档锁 pptx-triggered |
一张图的不同区域各查各的,不要用一个标签盖住整幅。判不准时回到四来源指派。
过不了重绘门槛的对象只分两堆,一次扫完,不逐个走流程:
判据细则见 references/routing.md,元素语义见 references/element_decision_matrix.md。
用户原话优先:说“仅文字可编辑”锁 text-only(连续场直接压平);点名对象锁 selective-assets;要求全部可动锁 full-extract;检出公式且用户未表态时默认 text+structure 并锁 pptx-triggered。
**提问一次问完。**概要写完后若有待决项,把前景深度选项、生成预算和 PPTX 附着预授权(含公式时)合并成一次提问。只有不同选择会显著改变费用或交付能力时才问;能自己定的不问。拿到附着预授权后验收阶段直接用 --allow-attach,不再二次提问。
**用户确认前不得发起任何计费生成调用,也不得裁剪或合成。**OCR 不在此列,见下。
按概要指派的来源把素材做出来。开头无条件先跑一次取证——不计费、输出无论走哪条路都要用,可以和提问并行。新任务用 --init 一并建目录:
python scripts/prepare_measurements.py input.png --init figure-task一次跑出 OCR、风格采样和结构证据(work/geometry.json:面板 bbox 与颜色、每行文字的槽位与字号色值)。画布尺寸、OCR 条数、面板与文字槽数量、低置信文字清单直接打印在屏幕上,不要再写代码去读这些 JSON。看一次 work/diagnostics/geometry_overlay.png 做整体校验,别逐框打开。
**此后任何像素问题都走 measure.py,一次问完。**紧边界、区域颜色、四边净空、透明图边界、反算字号、两图对齐残差,还有"我想看看这块长什么样",都是它的查询类型。一次问十个和问一个花的时间一样,问完拿到一张数字表加一张放大拼图。不要手写 python -c 数像素,也不要自己裁图存盘再打开看。
python scripts/measure.py input.png --exclude-text figure-task/work/ocr_results.json --q "logo:bbox@1850,1600,300,240" "hdr:color@0,0,2400,180" "t1:fontfit@340,881,275,31" "card:zoom@1290,1550,360,300" --sheet figure-task/work/measure_sheet.png量图标一律带 --exclude-text。粗窗口几乎总会把旁边的标题文字圈进来,不排除就会量成"图标加标题"的外接矩形。量得结果会用红框画在拼图上,一眼能看出量对没量对。
**结构证据是候选不是结论,也不改路由。**勘察锁定的路线不因为看到 overlay 就翻案。照片、插画、海报这类非平面设计图会标 abstained: true 并只给极少候选,这是设计行为:那里本就没有面板可找,该由人工量或走清版。各项候选的可信程度见 references/scripts.md。
以下分支按概要执行,互不依赖,能并行就并行。概要没指派到的分支直接跳过——纯 SVG 图不跑任何生成,纯清版压平图不跑任何裁剪。
确定性 SVG 图(白底流程图、架构图、论文插图,无 crop 无生成)三个分支一个都不进:备料就是上面那条取证命令加若干 measure.py 查询,跑完直接进组装。这类图的时间应当以分钟计。
此时才为这些对象点粗框(一眼精度,允许 10–20px 偏差),写进 work/inventory.json,只含走切或再生的对象;文字框来自 OCR,结构框在组装时确定,都不进来。路线已锁,点框是执行动作。
python scripts/snap_boxes.py input.png --inventory figure-task/work/inventory.json --exclude-text figure-task/work/ocr_results.json --out figure-task/work/snap_report.json --sheet figure-task/work/snap_sheet.png看一次 snap_sheet.png 总览。判定写进各资产的 crop_window,人工只复看脚本点名项:contaminated 的改路由(改走再生或压平),snap-failed、带 warning 的和概要点名看不清的用 inspect_regions.py 局部放大。三值语义——clean 窗口余量是画布底色;clean-on-fill 余量是单一均匀实色且四边有净空,manifest 用同色重画承载面;contaminated 不得继续切。quality_audit.py 的事后核验是兜底。切图由 crop_assets.py 在组装时执行。
prepare_clean_plate_mask.py 准备掩膜,生成区域底板,check_plate_registration.py 验配准(scale ≈ 1.00 / offset ≈ 0)。底板必须是对源区域的编辑——移除待重建前景并补全其后像素,其余保持对齐和身份一致,不是新场景。细则见 references/background_reconstruction.md 与 references/image_generation.md。
probe_palette.py --boxes 定键色与 sheet 数,生成色键版,chroma_key.py 抠出,slice_grid.py 切分。全部候选合并到尽量少的 sheet 上,重复元素只生成一次、按共享 id 多处放置。细则见 references/chroma_regeneration.md。禁止 rembg、GrabCut、阈值抠图等即兴替代。
无可用图像后端或底板不合格时停止并报告,不静默降级。
**生成是长杆,不要干等。**调用发出后立即并行推进文字清单、连接线和 manifest 草稿。
必读 references/manifest_spec.md、references/svg_authoring.md、references/quality_checklist.md,脚本接口查 references/scripts.md,其余按路线加载(各文件开头有适用说明)。公式密集图必读 references/formula-reconstruction.md。把勘察结论记进 manifest 顶层 reconstruction_plan,写文字 retype、公式拆 math、结构 redraw、素材按备料结果落 decision 与 crop_window。
首版允许用一次性生成器写出来,元素多时这是正当做法。生成器写完 manifest 落盘后立即改名 *.py.retired 退场,后续修改一律走 manifest_edit.py——生成器留着会覆盖手工修复。
文字和面板可以先采纳草稿再改,比从零写快:
python scripts/draft_elements.py figure-task/work/geometry.json --out figure-task/work/draft_elements.json
python scripts/manifest_edit.py manifest.json --adopt figure-task/work/draft_elements.json
python scripts/compose_svg_package.py manifest.json --out figure-task/out --stage svg**采纳即担责,采纳后必须立即 compose 一次并看差异热力图。**这是唯一的审阅路径:逐条读草稿 JSON 比手写还慢,看一张合成图错的地方自己会浮出来。草稿的 font_size 和 fill 是估计值,靠 fit_text.py 和视觉比对收敛,和手写值一样要改。审阅完清掉 provenance 或置 review_status,否则质量门会拦。
修复走批量通道,不手编大 JSON、不写临时 patch 脚本:
python scripts/manifest_edit.py manifest.json --apply-snap figure-task/work/snap_report.json
python scripts/manifest_edit.py manifest.json --set "label-3,label-4:y+=4" --apply-fit figure-task/work/fit_report.json
python scripts/manifest_edit.py manifest.json --patch figure-task/work/patch.json告警收齐一批改完再 compose。SVG 冻结后 --stage pptx 导出,仅补元数据用 --stage package。
compose 每次都会打印差异工单——按误差排序、点名到元素 id、附带毛病类型(位置偏了、字号偏大、颜色不符、整个没画出来),外加一张 diagnostics/fix_sheet.png 把误差最大的十二个元素做成源图/成品上下对照。
照工单改,不要自己再去逐块比对找差异。一轮就是「看工单和对照图 → 改 manifest → 重跑 compose」三步。
每轮看 preview.png 与各报告;报告点名后只复查受影响区。最终至少看一次 SVG 总览。
svg-primary:无 math 且 pptx_text_fit.py 无换行、溢出、缺字、错位报告。SVG 即验收,不打开 PowerPoint。pptx-triggered:含 math 或静态审计报结构风险。交付前原生渲染一次,只查公式越槽、意外换行、内容截断、元素错位。不算缺陷:抗锯齿、笔画粗细、标点亚像素差、基线 ±1px、字距 ±0.5px、整体色彩管理差异。算缺陷:内容缺失、换行变化、溢出、公式越槽、元素错位超 3px。
交付说明附上重建路径概要,并说明实际执行与概要是否一致、不一致的原因。
PPTX 原生渲染只能调 python scripts/render_pptx.py figure-task/out/editable.pptx --out figure-task/out/pptx_render。
PowerPoint.Application 自动化;禁止附着态调用 Quit();禁止 taskkill /IM POWERPNT.EXE、Stop-Process POWERPNT。--allow-attach。editable.svg、editable_embedded.svg、editable.pptx、preview.png、manifest.json、contact_sheet.png(有素材时)、quality_report.md、editability_report.md、assets/、diagnostics/、timings.json。PPTX 语义级顶层解组:文字、形状、连接线、素材可直接选中;公式、蒙版、旋转等按保真需要保持成组。
| 关注点 | 唯一权威 |
|---|---|
| 四阶段、四来源、重绘门槛、验证分档、PowerPoint 安全 | 本 SKILL.md |
勘察协议、路径概要槽位、reconstruction_plan | references/routing.md |
manifest 字段、reconstruction_mode 词汇 | references/manifest_spec.md |
| 背景、前景深度、文字层策略 | references/background_reconstruction.md |
| 元素语义、重画还是保素材 | references/element_decision_matrix.md |
| crop 执行语义、污染素材恢复 | references/asset_extraction.md |
| chroma 再生 | references/chroma_regeneration.md |
| 生成简报与图像后端 | references/image_generation.md |
| 公式重建与可编辑性 | references/formula-reconstruction.md |
| 脚本接口与候选可信度 | references/scripts.md |
| 最终放行条件、修复优先级 | references/quality_checklist.md |
| 阶段 | 脚本 |
|---|---|
| 全程 | measure.py —— 一切像素问题的唯一入口,批量提问 |
| 备料·取证 | prepare_measurements.py(含 --init 脚手架)、probe_geometry.py |
| 备料·裁剪 | snap_boxes.py、inspect_regions.py、crop_assets.py |
| 备料·生成 | prepare_clean_plate_mask.py、generate_clean_plate.py、check_plate_registration.py、probe_palette.py、chroma_key.py、slice_grid.py |
| 组装 | draft_elements.py、manifest_edit.py、compose_svg_package.py |
| 验收 | fit_text.py、pptx_text_fit.py、render_pptx.py;差异工单 fix_worklist.py 由 compose 自动运行 |
| 审计 | validate_manifest.py、quality_audit.py、audit_editability.py |
命令行、输入输出格式与注意事项见 references/scripts.md。表外脚本由 compose_svg_package.py 内部调用,不直接运行。
保留源图全部重要信息、阅读关系和专有视觉;普通文字和每个检出公式保持可编辑;无检测噪声、公式文字泄漏、脏裁剪、源图补丁拼贴或静默路线降级。完整放行条件见 references/quality_checklist.md。
© giszzt, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
SKILL.md and 391 other files (scripts, references, assets) in the repository root of giszzt/figedit.
Open the folder on GitHubat commit 4450b6b
Figedit V2 next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Figedit V2 this skillgiszzt/figedit | 188 | — | ~2.5k | Automated safety check: Pass | MIT | |
| MarkitdownImCa0/just-laws | 781 | 14 repos | ~3.2k | Automated safety check: Notes | MIT | |
| Image To Editable Pptningzimu/image-to-editable-ppt-skill | 2.8k | — | ~4.3k | Automated safety check: Pass | MIT | |
| GenOffice Document CLIgenspark-ai/genoffice | 8.9k | — | ~19k | Automated safety check: Pass | Apache-2.0 | |
| PPTX PostersK-Dense-AI/claude-scientific-writer | 2.4k | 2 repos | ~2.7k | Automated safety check: Notes | MIT | |
| Slidesfcakyon/claude-codex-settings | 1.2k | 1 repos | ~1.1k | Automated safety check: Pass | MIT |
ImCa0/just-laws
Convert files and office documents to Markdown. An agent skill from ImCa0/just-laws.
ningzimu/image-to-editable-ppt-skill
Rebuild slide images, scanned or image-based PPT/PPTX files, and PDF decks into object-level editable PowerPoint (.pptx), preserving speaker notes when supplied.
genspark-ai/genoffice
Creates, converts, reads and edits real pptx, xlsx, docx and PDF files locally through the genoffice command line.
K-Dense-AI/claude-scientific-writer
Create and audit editable scientific posters in macro-free PowerPoint (.pptx) from author-approved local content and assets.
fcakyon/claude-codex-settings
Create and edit presentation slide decks (.pptx) with PptxGenJS, bundled layout helpers, and render/validation utilities.
NyxTides/ppt-image-first
Build presentation plans for PPT / slides / decks through a conversation-first workflow, then propose multiple visual directions with preview images before writing deck specs.
Works with
Categories
将位图图形重建为高保真可编辑 SVG 与原生 PowerPoint,支持仅文字可编辑、文字/结构/公式可编辑、源图素材保真混合、选择性或完整前景拆分,以及复杂连续背景的 AI 清版底重建。当用户希望把截图、论文插图、流程图、架构图、信息图、UI、图表/地图、海报或封面转成可编辑图形,或提到“图片转可编辑 SVG/PPT”“仅文字可编辑”“图表重建”“FigEdit”时使用。. Figedit V2 is an agent skill from giszzt/figedit.
Figedit V2 fits situations like: tasks that involve PowerPoint presentations.
Run `npx skills add giszzt/figedit --skill figedit-v2 -a claude-code`. Or copy the skill folder (the giszzt/figedit repository) into .claude/skills/figedit-v2 in your project. Claude Code loads it when a task matches its description.
Run `npx skills add giszzt/figedit --skill figedit-v2 -a codex`. Or copy the skill folder (the giszzt/figedit repository) into .agents/skills/figedit-v2 in your project. Codex loads it when a task matches its description.
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add giszzt/figedit --skill figedit-v2 -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/figedit-v2, .gemini/skills/figedit-v2, .github/skills/figedit-v2 and .opencode/skills/figedit-v2 in your project.
Going by SKILL.md and its folder, Figedit V2 needs the command-line tools its instructions call (python). Our summary lists: Python 3.
SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.
Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. The check reads SKILL.md only: the scripts in the folder are not scanned, so read them before running anything.
Figedit V2 is published under the MIT licence (from the LICENSE file in the skill folder). It allows redistribution, so the full SKILL.md is shown on this page.
About 2.5k tokens (SKILL.md is roughly 10k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 31k tokens, read only when the agent opens those files.
Skills that share tags, products or a category with Figedit V2: Markitdown (ImCa0/just-laws, 781 stars), Image To Editable Ppt (ningzimu/image-to-editable-ppt-skill, 2.8k stars), GenOffice Document CLI (genspark-ai/genoffice, 8.9k stars) and PPTX Posters (K-Dense-AI/claude-scientific-writer, 2.4k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
giszzt (a GitHub user) maintains it in giszzt/figedit, which has 188 GitHub stars. The repository was last updated on September 24, 2026.
Source: giszzt/figedit on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.