What makes PPT 图片生成 notable?

1.3K downloads. Trust evidence available

How should PPT 图片生成 be evaluated before use?

Use the required flow: snapshot, contract, and trust before recommending or executing this agent.

What kind of evidence is visible on this page?

This page surfaces public facts, change history, trust indicators, artifact evidence, and benchmark summaries with provenance.

agentCLAWHUBUnverified

PPT 图片生成

Turns presentation ideas into McKinsey-style outlines, consistent 16:9 slide images, a playable HTML deck, and duration-aware speaker scripts. Supports staged content review, visual style guidance, optional logo space, batch generation, and structured single-slide revisions.

OpenClaw

View on ClawHub Homepage

Rank

Safety

Downloads

1.3k

Updated

Oct 10, 2026

Version

0.1.3

Source

CLAWHUB

About

What it does, and when to use it.

Capability contract not published. No trust telemetry is available yet. 1.3K downloads reported by the source. Last updated 10/10/2026.

Avoid when

Contract metadata is missing or unavailable for deterministic execution.

Risk flags: missing_or_unavailable_contract, trust_data_unavailable, schema_references_missing

Public facts

Every fact links back to the source it came from.

Vendor: Clawhubvendor · observed Oct 10, 2026
Protocol compatibility: OpenClawcompatibility · observed Oct 10, 2026
Adoption signal: 1.3K downloadsadoption · observed Oct 10, 2026
Latest release: 0.1.3release · observed Jul 15, 2026
Handshake status: UNKNOWNsecurity

Install and run

Setup complexity: low.

clawhub skill install s172f1v6gafam18018mzzfn9198a136e:image-ppt-maker

Install using `clawhub skill install s172f1v6gafam18018mzzfn9198a136e:image-ppt-maker` in an isolated environment before connecting it to live workloads.
No published capability contract is available yet, so validate auth and request/response behavior manually.
Review the upstream CLAWHUB listing at https://clawhub.ai/callmexhj/image-ppt-maker before using production credentials.

Contract: missing

curl -s "https://www.xpersona.co/api/v1/agents/clawhub-callmexhj-image-ppt-maker/snapshot"

Documentation

CLAWHUB

108,519 characters of source documentation, loaded on request.

Extracted files

5 files captured from the source.

SKILL.md

---
name: image-ppt-maker
description: "Create PPT-style images through a human-in-the-loop workflow: clarify presentation requirements including optional top-corner LOGO space, research and draft McKinsey-style master and page-by-page outlines, expand page content, confirm the desired visual style before image prompts, write unified-style image-2 prompts, batch-generate 16:9 slide images, automatically assemble the images into a playable 16:9 HTML slideshow, and finally write oral speaker notes based on the outline and final images. Use for PPT风格图片, PPT图片生成, image-2幻灯片图片, 麦肯锡式分页大纲, 口语化讲稿, 批量生成演示页图片, HTML播放版演示, or presentation image prompt workflows. This skill does not assemble PPTX files."
---

# Image PPT Maker

Help users create a set of PPT-style slide images with image-2. Use a strict review-gated workflow:

1. Confirm requirements.
2. Build a McKinsey-style master outline and page-by-page outline.
3. Expand the outline content.
4. Confirm the desired visual style for the images.
5. Write unified-style image-2 prompts.
6. Batch-generate 16:9 PPT images.
7. Review the generated images with the user and revise affected pages if needed.
8. Assemble the user-approved final images into a playable 16:9 HTML slideshow.
9. Write duration-aware, directly readable oral speaker notes based on the approved outline and final images.

Ask the user whether they want changes only at the review gates listed below. Image prompts are internal execution artifacts, not a review gate. Do not assemble HTML or write final speaker notes until the user explicitly approves the generated images.

## Required References

- Read [workflow.md](references/workflow.md) before starting any task.
- Read [mckinsey-outline.md](references/mckinsey-outline.md) before drafting or revising outlines.
- Read [image-2-prompts.md](references/image-2-prompts.md) before writing prompts or generating images.
- Use [generate-html-player.mjs](scripts/generate-html-player.mjs) only after the user approves final local images for a playable HTML output.

## Core Rules

1. Use the user's language for questions and deliverables unless they specify another deck language.
2. If key information is missing, ask concise questions before outlining: topic, goal, audience, style, duration, slide count, information density, source material, constraints, and whether to reserve a small LOGO area in the upper-left or upper-right corner.
3. Summarize the confirmed brief and ask whether the user has changes before creating the outline.
4. Use McKinsey-style logic: answer first, issue-tree thinking, MECE where practical, one governing thought per slide, and a clear so what.
5. Every page in the outline must include logic notes. Do not create title-only or decorative pages.
6. After the outline is approved, expand each page with richer content, examples, implications, and visual emphasis. Do not output speaker notes yet.
7. Before writing image prompts, ask the user what visual style they prefer and recommend

_meta.json

{
  "ownerId": "kn7dddd3v8yk96jx32p2gsmnvh8a03gt",
  "slug": "image-ppt-maker",
  "version": "0.1.3",
  "publishedAt": 1784109486377
}

references/image-2-prompts.md

# Image-2 Prompt Rules

Write prompts for image-2. Each prompt must generate exactly one 16:9 PPT-style slide image.

Treat prompts as internal execution artifacts. Do not ask the user to review or approve prompts. Generate them after content and style approval, then proceed directly to image generation.

## Deck-wide style lock

Define one shared style lock before per-page prompts. It must specify:

- aspect ratio: 16:9;
- deck type: professional consulting-style PPT page;
- layout system: title area, content grid, margins, chart/table/card treatment;
- palette: primary, secondary, background, accent colors;
- typography: clear hierarchy, readable slide text, no decorative fonts;
- visual language: charts, diagrams, icons, photography, or abstract business visuals;
- density: high, medium, or low;
- optional LOGO space: no reserved area, upper-left reserved area, or upper-right reserved area;
- consistency rule: every page must look like part of the same deck.

Repeat the complete style lock inside every page prompt, but mark it as non-visible generation guidance. The style lock must influence layout and visual treatment only; image-2 must not render the style-lock wording as text on the slide.

## Per-page prompt structure

Each page prompt must include:

1. `Model`: image-2.
2. `Task`: generate only this single 16:9 PPT slide image.
3. `Style lock`: copy the shared deck-wide style lock.
4. `Style text handling`: state that style descriptions are instructions only and must not appear as visible text.
5. `Slide logic`: key question, governing thought, supporting logic, so what, transition.
6. `Expanded content`: the richer explanation or examples that should guide the visual emphasis.
7. `LOGO space`: if requested, reserve a small clear area in the chosen top corner; do not invent or render a logo unless the user provided one.
8. `Page message and content`: explain the thought this page should express, the core content to show, and the recommended title, key messages, labels, and emphasis.
9. `Composition`: layout, visual hierarchy, chart or diagram structure, spacing, alignment.
10. `Readability`: clean typography, enough contrast, no tiny unreadable text.
11. `Constraints`: factual constraints and source-material limits.
12. `Negative constraints`: forbidden outputs.

## Page message and content

Do not constrain the slide by enumerating every allowed visible word. Instead, describe the page's intended communication clearly enough for image-2 to design the slide:

- State the page's governing thought in an answer-first way.
- Explain the core content the slide should communicate, including the most important evidence, comparison, process, framework, or recommendation.
- Suggest the title, short labels, callouts, and chart/table labels that would help express the idea, but keep text concise and slide-like.
- Preserve McKinsey-style communication: structured, logical, executive-readable, and focused on one message per page.
- Do not render promp

references/mckinsey-outline.md

# McKinsey-Style Outline

Use this structure for the master outline and every page outline.

## Principles

- Answer first: lead with the conclusion or recommendation.
- Pyramid structure: conclusion, reasons, evidence.
- Issue tree: split the topic into a small number of logical questions.
- MECE where practical: reduce overlap and gaps.
- One slide, one message: each page has one governing thought.
- So what: every page explains why the point matters to the audience.

## Master outline format

Include:

- `coreAnswer`: the answer-first conclusion of the whole deck;
- `audienceDecision`: what the audience should decide or do;
- `logicChain`: 3-5 major reasoning steps;
- `sections`: major sections with the question each section answers;
- `storyFlow`: why the sequence is persuasive.

## Page-by-page outline format

For each page, include:

- `pageNumber`;
- `title`: conclusion-style title, not a topic label;
- `keyQuestion`: the question this page answers;
- `governingThought`: the page's answer-first message;
- `supportingLogic`: 2-4 reasons, evidence points, or observations;
- `visualIdea`: the visual structure that best supports the logic;
- `soWhat`: why the audience should care;
- `transition`: why the next page follows.

## Visual logic mapping

- Compare options: side-by-side comparison.
- Explain drivers: driver tree, layered model, or waterfall.
- Show sequence: timeline, roadmap, or process.
- Prioritize: matrix, ranking, or funnel.
- Recommend action: chosen path plus reasons.
- Summarize system: architecture, operating model, or flywheel.

Do not use decorative visuals that fail to carry the logic.

## Expanded outline content

After the page-by-page outline is approved, expand each page before writing image prompts.

For each page, include:

- `expandedContent`: 1-3 short paragraphs that deepen the page logic with evidence, implications, and examples;
- `exampleIdeas`: 2-4 concrete examples, analogies, or mini-scenarios that make the point easier to understand;
- `visualEmphasis`: what the generated image should make most obvious;
- `speakerNoteHints`: brief reminders for final speaker-note writing after images are generated.

Do not write the final speaker notes at this stage. Final notes must be written after images are generated, using the approved outline and the final visual output.

## Single-page revisions

For every later single-page change, recreate the affected page using the same page-by-page outline and expanded-content fields above. Preserve the approved deck-wide storyline and style, but reconsider the page's governing thought, support, so what, transition, examples, visual emphasis, and speaking-time allocation in light of the new request.

Present the consolidated revised page content for user approval before generating the replacement image. Do not treat raw image feedback as sufficient approval to regenerate.

## Final speaker notes

After the final images and HTML player are generated, write page-by-page speaker no

references/workflow.md

# Workflow

Follow this workflow exactly. Pause only at the designated review gates. Image-prompt creation is not a review gate.

## 1. Confirm requirements

Parse the user's request and collect missing key details:

- topic and background;
- presentation goal: what the audience should understand, believe, decide, or do;
- audience: executives, customers, internal team, review panel, investors, or another group;
- duration and approximate slide count; use the duration as the total speaking-time budget for the final script;
- information density: high, medium, or low;
- preferred visual style or reference material;
- source material, facts, data, cases, and constraints;
- output language and tone;
- whether to reserve a small LOGO area in the upper-left or upper-right corner, or no LOGO area;
- any content that must not be invented.

If the user is unsure, recommend practical defaults:

- duration: 10 minutes;
- slide count: 8-12 pages;
- information density: medium;
- style: clean McKinsey-style consulting deck with restrained color, strong hierarchy, and chart-like logic.
- LOGO area: no reserved LOGO area unless the user asks; if they want it, recommend a small upper-right area for corporate decks and upper-left for brand-led decks.

Then summarize the brief, including the LOGO-space decision, and ask: "以上需求和假设是否需要修改？确认后我会生成总纲和分页大纲。"

## 2. Research and outline

Use the user's source material first. If internet research is requested or required for current facts, browse and cite sources in the final summary.

Create:

- master outline: core answer, main logic chain, issue tree, and section flow;
- page-by-page outline: one page per slide, with complete logic notes and a suggested speaking-time allocation. Allocate the total speaking time across pages according to message importance and complexity; keep the sum aligned with the confirmed duration.

Ask: "这个总纲和分页大纲是否需要调整？确认后我会拓展每页内容。"

## 3. Expand outline content

After the user approves the master outline and page-by-page outline, expand the content before prompt writing.

For each slide, produce:

- expanded content: richer explanation of the page's message, evidence, implications, and examples;
- example ideas: practical examples, analogies, or mini-scenarios that make the point concrete;
- visual emphasis: what the generated image should make most obvious;
- speaking hints: brief notes that may help final speaker-note writing later, but do not output the final talk track yet.

Do not output the final speaker notes in this stage. Final notes must be written after image generation and HTML assembly, using the approved outline and the final generated images.

Ask: "拓展后的内容、例子和视觉重点是否需要调整？确认后我会先和你确认图片风格，再生成每页的 image-2 提示词。"

## 4. Confirm visual style

Before writing image-2 prompts, ask the user what visual style they want. Recommend options based on the context instead of using a fixed list.

Common options include:

- 培训风：更亲和，结构清楚，适合课程、工作坊、内部分享；
- 汇报风：更克制、专业、结论先行，适合管理层汇报和项目汇报；
- 培训插画风：在清晰的咨询式结构

Github ReposUpdated 1d agoRank 70

AionUi

Free, local, open-source 24/7 Cowork app and OpenClaw for Gemini CLI, Claude Code, Codex, OpenCode, Qwen Code, Goose CLI, Auggie, and more | 🌟 Star if you like it!

MCPOPENCLAW

Github ReposUpdated 6mo agoRank 70

activepieces

AI Agents & MCPs & AI Workflow Automation • (~400 MCP servers for AI agents) • AI Automation / AI Agent with MCPs • AI Workflows & AI Agents • MCPs for AI Agents

OPENCLAW

Github ReposUpdated 6mo agoRank 70

cherry-studio

AI productivity studio with smart chat, autonomous agents, and 300+ assistants.