Pitchcraft logo

Pitchcraft

Community
huyanxius
pitchcraft

路演构造术:从产品仓库、GitHub URL、讲稿或既有 PPT 提炼证据与产品视觉风格,用 RW 生成整页图片初稿,再转换为保留版式的可编辑 PPTX。用于比赛答辩、产品路演,以及“rwppt 初稿转可编辑”的完整流程;支持从已有图片稿或部分可编辑稿续做,并按需交付讲稿、时间表、Demo 预案与评委问答。

Overview

Publisherhuyanxius
Repositorypitchcraft
Skill namepitchcraft
Stars
4
Forks
0
Bundled files
9
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 9 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by huyanxius on GitHub. Read the source before you install it.

Installation

Install the Pitchcraft AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/huyanxius/pitchcraft/tree/main/skills/pitchcraft
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/huyanxius/pitchcraft.git /tmp/pitchcraft
mkdir -p .claude/skills
cp -r /tmp/pitchcraft/skills/pitchcraft .claude/skills/pitchcraft
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Pitchcraft in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Pitchcraft on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Pitchcraft is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Pitchcraft|路演构造术

把真实产品和可靠材料组织成能讲清楚、看得懂、方便修改的路演。默认中文、比赛答辩场景、产品自身的视觉风格。共同优化速度、内容质量、视觉质量和修改成本,不以更多生成次数或更多流程文档代表质量。

一套入口,两个制作阶段

  1. 图片初稿:使用 rw-consulting-ppt 完成结论、证据主体、视觉层级与完整页面生成。每页一张 PNG,图片版 PPTX 每页恰好一张图。
  2. 可编辑定稿:使用 image-ppt-to-editable 将认可的页面恢复为可见的原生文字、适当的原生图形与有意义的独立素材;底图不得重复留字。最终调用当前 Presentations 技能的组装与验证能力。

这两个契约分别作用于不同文件。RW 的“零可编辑文本”只检查图片初稿,不限制后续转换;转换阶段不得借机重新设计认可的初稿。图片版是可直接使用的正式交付,并非必须继续转换。明确提供两种终点:图片 PPT,或图片 PPT 加可编辑 PPT。按用户选择执行;未说明时在首次简短对齐中确认一次,不默认为必须制作两版。用户只要图片、样张或检查时按其范围停止。

本 skill 负责编排、事实和路演表达,复用两个下游技能,不复制整套提示词、抠图脚本或 PPT 组装器。开始相关阶段时,按当前技能目录找到并读取下游 SKILL.md,再加载它要求的相关引用;不要硬编码某次插件版本路径。所需技能或生成能力缺失时,说明缺失项,完成不依赖它的工作,但不能以不同格式冒充最终交付。

进入点与授权

当前输入从哪里开始
仓库、资料或讲稿,尚无成稿核对证据、故事线、图片样张,再完成两阶段
已有认可的整页 PNG、图片 PPT/PDF直接转换,复用原文、素材与风格,不重新生成初稿
已有文本框和拆好的图片,仍有漏字或残字直接走 image-ppt-to-editable 的既有 PPT 修复;图片对象不重新拆分或合并
用户只改某页、只试样张或只做检查限定到指定范围,不自动扩成整套重做

以用户最新确认的文件、内容与风格为准。进场先清点已有产物和实际完成状态,保留源文件并另存。已有批准、已知时长和已选风格持续有效,不重复询问。

工作节奏

  • Standard(默认):将受众、时长、故事线和样页选择压缩成一次简短对齐;生成 1–2 张代表样页,按 RW 的样张规则等待认可后批量。已明确认可的故事线或样张不再审批一次。
  • Emergency:用户明确要求“不要停、一口气完成”时,直接执行其已授权范围,减少外部确认;仍完成内部事实、整页和文件检查。仅说“快”不等于取消用户已要求的样张确认。
  • Final:在制作基础上补证据核对、真实计时排练、转场、答辩和 Demo 失败预案。没有实际计时,只能称为预计时间,不能声称排练通过。

不要为了遵循模式替用户增加待办。用户已要求完整可编辑交付时,图片初稿认可即进入转换;只有转换样张存在实质性取舍或用户明确要求再看时才额外等待。

1. 核对内容、产品与视觉来源

仓库输入先读项目 AGENTS.md、HANDOFF.md、TASKS.md,保持只读扫描;读 产品与视觉来源。需要扫描时运行本 skill 的脚本:

bash
python3 <skill-dir>/scripts/scan_product_repo.py <local-path-or-github-url> --out <work-dir>/repo-scan

脚本支持仓库 URL 与 PR URL;默认分支可能不代表产品。用户指定分支、版本或文件优先,脚本计数和颜色频率只作线索。资料型输入不强求仓库,不为做 PPT 新建 Git 项目。

以一份简短记录确定:产品与用户问题、可演示流程、最强证据、确认文案、来源与缺口、认可素材、配色/字体/材质/密度、排除的旧 Demo 内容。区分已演示、已实现、文档声称、计划、假设与历史材料,不把测试通过当真实用户成效。

沿用产品风格或用户认可样页的整页节奏,不只套 Logo 和颜色。真实界面素材用于证明实际流程;示意图不能冒充产品运行证据。使用已提供或已有合法素材,不截图、不上传截图。

2. 建立定时故事线与逐页正文

读 故事线与时间。只补问真正影响结果且无法从材料获得的信息:受众、时长、期望判断和交付范围。仅转换既有 PPT 时无需重新询问路演时长。

把方向卡、正文与逐页 brief 合并维护,避免多份近义文档反复对齐。每页记录:

  • 一个完整结论与它在讲述中的作用,标题不喊空口号。
  • 一个能支持结论的主要证据图、过程、案例或对比。
  • 必须保留的准确文字、姓名、数字、单位、网址和来源。
  • 证据状态与不能暗示的结论。
  • 视觉方向、资产引用、演讲秒数和下一页转场。

正文先定再生成。首次转换优先用已确认正文与坐标核对,OCR 只补定位和发现遗漏;不要丢掉上游正文后从图片重新猜一遍。

3. RW 图片初稿

进入 rw-consulting-ppt,读 本技能的生成边界,按需加载 RW 的视觉治理引用。输入已确定的正文、证据和产品风格,沿用它们,不再次问一轮相同问题。

通过原生 imagegen 生成完整 16:9 页面,文字也在该页面图中;禁止用 HTML、代码绘图或原生 PPT 排版栅格化来替代图片初稿。用代表样页验证标题层级、证据主体和文字密度,认可后锁定整套风格。

稳定命名为 slide_01.png 等,逐页核对必保文字、证据含义与实际比例;局部问题只改受影响页面。必要的图中文字纠错应发生在认可图片初稿之前,不能默许图片版有错而只在可编辑版中偷偷修正。

用既有脚本机械打包图片版,不编写另一套打包器:

bash
python3 <skill-dir>/scripts/package_image_deck.py <slide-image-dir> --out <output-dir>/roadshow-image-draft.pptx

4. 从认可初稿转换为可编辑稿

转换前必须读 两阶段交接与续做 和 image-ppt-to-editable。先完成一张代表页的底图清理、文字恢复、组装与整页检查,再按页或短批次推进;不积累大量素材而迟迟没有完整页面。

原初稿为整页位图时,以有意义的完整视觉对象划定素材边界,避免文字一块、箭头一块的碎切;已拆好的 PPT 则保持原图片数量、位置、尺寸、裁切、旋转和组织关系。两种情况不可混淆。

复用原素材和已经可用的去字成果。页脚、空白条和可复用底纹不默认单独生成。检查生成图是否新增留白或改变比例,检查跨图文本、中文字形度量、旋转、文字颜色、真实透明度与照片/界面保护范围。

图片素材里的目标文字必须真正清掉;原生文字应可见可编辑,不能用隐藏文本或遮盖脏底图虚报完成。复杂照片可保留为位图,图表只有源数据可靠才称数据可编辑。

5. 检查并交付

读 输出与验收。分别检查图片版和可编辑版,不对两个版本使用同一组对象数量规则。批量前验证中文渲染器;结构通过后再看完整页面。只复查修复所影响的页面与共享资源引用,不为了更保险反复全量重做。

路演按所选终点交付图片 PPT 或再加可编辑 PPT;完整路演按需附讲稿/时间表、Demo 预案、评委问答和简短来源说明。用户仅要 PPT 或转换时不强制扩展附加材料。

用户询问进度时直接报已回装且检查的页数,生成图片数另列。以整页和可打开的 PPT 展示成果,少输出细碎过程图和重复承诺。已认可页面冻结,全部约定问题解决、必要检查通过并另存后停止。

调用示例

  • 用 $pitchcraft,根据这个仓库做 8 分钟答辩:先 RW 图片初稿,再给可编辑 PPT。
  • 用 $pitchcraft,沿用这些已认可 PNG,直接转可编辑,不重新生成。
  • 用 $pitchcraft,只清这份已拆好 PPT 的残字并补可编辑文字,保持图片组织。
  • 用 $pitchcraft,只做前两张图片样稿,等我看过再继续。

速度与局部修改的补充判断见 执行经验。

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Pitchcraft AI skill do?

路演构造术:从产品仓库、GitHub URL、讲稿或既有 PPT 提炼证据与产品视觉风格,用 RW 生成整页图片初稿,再转换为保留版式的可编辑 PPTX。用于比赛答辩、产品路演,以及“rwppt 初稿转可编辑”的完整流程;支持从已有图片稿或部分可编辑稿续做,并按需交付讲稿、时间表、Demo 预案与评委问答。

Why use Pitchcraft on TypingMind?

Because you install it once and use it with any model. Pitchcraft is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Pitchcraft in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/huyanxius/pitchcraft/tree/main/skills/pitchcraft. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Pitchcraft?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Pitchcraft?

As many as you like. As long as a model supports skills, you can use Pitchcraft with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Pitchcraft AI skill free?

It is published on GitHub by huyanxius. Check the repository for licensing terms. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇