PPT Agent
Generate professional presentations the way you build software — from a one-line request straight to a validated PPTX file.
- Source repo
- sunbigfly/ppt-agent-skills
- Stars
- ★ 926
- Last updated
- 5mo ago
- License
- NOASSERTION
- Primary language
- Python
- FA score
- 51/100 · Major gaps
At a glance
- How it runs
- Works with
- Portable with changes
- Cost
- Free software; you pay for model usage
- Setup effort
- Medium · a few setup steps
- You'll need
- Typical use
- A founder needing a 15-page pitch deck fast can give a single line (e.g. "embodied intelligence trends in 2026, dark tech style") and receive dual-format PPTX plus a web preview
- Not a fit if
- Agent platforms without Skill loading support need extra adaptation
- Teams unwilling to bear the high runtime and token cost of per-page screenshots and LLM visual audits
- Maintenance
- Commits in the last 6 monthslast commit 2026-06-08
- Source review
- 51/100 · Major gaps
What does this agent do, and when should you use it?
PPT Agent (sunbigfly/ppt-agent-skills) is a code-driven, state-machine-managed multi-agent framework that turns a one-line request into a professional PPTX file. It splits the pipeline into seven stages: interview, branch confirmation, web search or local source compression, narrative outline, global style lock, per-page parallel production (Planning → HTML → Visual QA), and dual-format export — each stage's artifacts are gate-checked before proceeding. Research / Outline / Style / Planning run as isolated subagents with non-overlapping contexts, and every subagent is forced to carry a SUBAGENT_MODEL parameter. Each page first produces a JSON contract validated by planning_validator.py, then renders to HTML, is screenshot automatically, and is audited pixel-by-pixel by an LLM, with overflow fixed by DOM + CSS structural rewriting. The built-in Puppeteer rasterization engine exports both PNG and SVG PPTX files, and all artifacts land in ppt-output/runs/<RUN_ID>/. Distributed as an Agent Skill with no separate deployment, it ships 8 themes, 10 layout types, 13 chart templates, and 15 scripts.
After the user submits a one-line request, the framework executes its state machine: P0 interview produces interview-qa.txt and requirements-interview.txt; P2A web search or P2B local source compression yields search.txt / source-brief.txt; P3 generates outline.txt and P3.5 locks style.json; P4 runs per-page parallel production — Planning (planningN.json validated by planning_validator.py) → HTML rendering (slide-N.html) → screenshot slide-N.png audited by LLM Visual QA, where detected overflow triggers DOM + CSS structural rewriting by a patch subagent; P5 builds preview.html and exports presentation-png.pptx and presentation-svg.pptx, finishing with delivery-manifest.. Execution relies on SKILL.md as the main console (state machine, gates, recovery rules), scripts in scripts/ (validators, harness, exporter), plus visual_qa.py (planning + html dual-layer assertions), subagent_logger.py, and planning_validator.py. Interrupted runs resume statelessly by scanning existing artifacts on disk (outline.txt / style.json / slide-N.png, etc.) to infer the recovery point.
- A founder needing a 15-page pitch deck fast can give a single line (e.g. "embodied intelligence trends in 2026, dark tech style") and receive dual-format PPTX plus a web preview
- A consultant who compresses local reports into decks can use the P2B local source compression branch to distill documents before per-page generation
- A content team sensitive to layout quality can rely on the pixel-level Visual QA loop to eliminate overflow, overlap, and hallucination issues
- An operations staffer producing structured decks with charts can reuse the 13 chart templates and 10 layout types directly
- A technical user who wants auditability can use subagent_logger.py stage commands and run logs to replay and redo from checkpoints
Key setup facts
Pulled from the install and usage notes below. The project README has the current values.
- Environment variables
RUN_ID
How do you install or deploy this agent?
Run in a terminal: npx skills add sunbigfly/ppt-agent-skills. The project runs as an Agent Skill and needs no standalone deployment; the README does not list system dependencies (Node.js version, browser requirements) beyond this install command, so verify against the repository scripts.
How do you use this agent?
After installing, enter a natural-language request in any Skill-capable agent environment, e.g.: "Generate a 15-page pitch deck on embodied intelligence trends in 2026, dark tech style." The framework runs all seven stages (P0–P5) automatically; all outputs (preview.html, presentation-png.pptx, presentation-svg.pptx, and intermediate contract files) are written to ppt-output/runs/<RUN_ID>/.
What are this agent's strengths and limitations?
- Subagent stage isolation keeps Research / Outline / Style / Planning contexts uncontaminated, and SUBAGENT_MODEL is mandatory per subagent to prevent silent fallback
- Pixel-level Visual QA loop: each page is screenshot and audited by an LLM; detected overflow is fixed via DOM + CSS structural rewriting rather than spacing tweaks, addressing root causes of overlap and layout chaos
- Data/render separation: JSON contracts pass planning_validator.py before driving HTML rendering, with an html-level physical structure detector blocking non-conforming skeletons
- Stateless checkpoint recovery: interrupts resume by scanning existing artifacts on disk; a failed step rolls back only that step, not other pages
- Dual-engine export: PNG raster stream guarantees cross-platform visual fidelity; SVG vector stream keeps fonts editable; the Puppeteer engine uses event hooks (document.fonts.ready) for hard-delay-free screenshots
- Requires an agent environment that supports Skill loading; platforms without Skill support need extra adaptation, and the README offers no standalone deployment path
- Subagents must have SUBAGENT_MODEL set with no default fallback, so users must configure model access and bear LLM call costs
- The multi-stage pipeline — per-page screenshots plus LLM visual audits — carries high execution time and token consumption
- The GitHub License field is NOASSERTION; although README badges say MIT, verify the LICENSE file before adopting
- Install docs omit concrete runtime requirements (Node.js version, browser dependencies); validate your environment before deployment
How does this agent compare with similar options?
Key facts side by side with the most closely related agents.
| Agent | Source review | Form / cost | Stars | Updated | Language | Full support on |
|---|---|---|---|---|---|---|
| PPT Agent This agent | 51 · Major gaps | Agent plugin / skillFree + model costs | ★ 926 | 5mo ago | Python | — |
| PPT Master — AI-native PowerPoint generation | 70 · Some gaps | Agent plugin / skillFree + model costs | ★ 79k | 1d ago | Python | Claude Code |
| OfficeCLI | 62 · Some gaps | CLIFree | ★ 33k | 1d ago | C# | Codex · Claude Code |
| Kineo (Kineo AI) | 34 · Major gaps | — | ★ 137 | 5d ago | TypeScript | — |
How does FollowAgents rate this agent?
Why each dimension lost points
SKILL.md requires a Step 0 interview, branch confirmation, manual audit gates, and isolated subagents; the artifact chain and per step outputs make data flow and rollback reasonably explicit. Deductions: these are mostly procedural requirements, with no concrete permission inventory, sensitive data rules, dependency pinning, or security review guidance. The scope of web research and script side effects is also not well bounded. The license names sunbigfly, but identity is unverified and no more specific source attribution policy is shown.
The documents define a fixed sequence, gates, validator rechecks, and retry or rollback behavior, giving the workflow reasonable internal structure; the README also describes artifact based recovery and visual QA. Deductions: the README's feature and performance claims are not corroborated by implementation evidence in this sample, dependency availability is lightly documented, and concrete failure messages or dependency failure handling are not shown.
SKILL.md covers Chinese and English presentation requests, document to deck conversion, pitches, training, and redesign of existing slides; it also describes web or local source branches, image modes, and manual review. Deductions: the trigger language includes broad implicit intents and may over-trigger; environment fit depends on CLI and subagent capabilities, but the sample provides no compatibility matrix.
The README has a clear workflow, artifact chain, repository map, quick start, example request, and update note, while SKILL.md is organized by stages. Deductions: installation guidance is mostly a single npx command, limitations and FAQ coverage are sparse, and although a version and date are shown, only one update entry is present. The license provides MIT text, but metadata says NOASSERTION; maintenance responsibility is not described beyond the copyright name, with no maintainer or update channel identified.
The product has a clear goal and deliverables, including HTML preview and PNG/SVG PPTX exports; staged planning, validation, and visual review indicate potential value beyond one shot generation. Deductions: quality, speed, and cross platform claims are mainly project assertions in the supplied material, with little comparative evidence, and the cost of the multi agent and rendering pipeline is not discussed.
The README and SKILL.md make workflow, artifacts, gates, and validation responsibilities traceable, and the license text is directly inspectable. Deductions: the sample contains no main scripts, validator, or tests to corroborate those descriptions; claims such as automatic screenshots, structural assertions, and checkpoint recovery remain documentation claims that cannot be checked against the supplied sources. The rationale therefore distinguishes explicit file content from uncorroborated outcome claims.
- The README's pixel level visual QA, automatic recovery, cross platform fidelity, and fast screenshot claims are not corroborated by implementation code in the supplied sample and should be treated as project claims, not verified capabilities.
- The workflow may search the web, read user supplied materials, and write HTML, images, and PPTX files to disk; users should review inputs and the output directory, since the sample does not explain boundaries for handling sensitive local content.
- The license file contains MIT terms, while the supplied license metadata is NOASSERTION; verify the repository's actual licensing status before publishing or redistributing.
FAQ
Do I need to deploy a standalone service?
What happens if generation is interrupted?
What guarantees layout quality?
What output formats are produced?
What is the license?
Is PPT Agent free?
Is PPT Agent safe to use?
Is PPT Agent still maintained?
Who makes PPT Agent?
Thanks — we use this to improve the page.