What does this skill do, and when should you use it?
deepPPT is an Agent Skill that generates academic PowerPoint decks for lab meetings, defenses, journal clubs, and conferences in Chinese or English. It turns error-prone layout steps into scripted checks: environment verification, PPTX text extraction, package-structure checks, layout QA, and PowerPoint/LibreOffice rendered previews. Its visual grammar (navy two-block title bars, segmented rules, chevron subtitles, opt-in red key-finding boxes) is fixed so decks from different agents look like one template. Runs are resumable, and the final PPTX is delivered only after both deterministic and visual QA gates pass.
- Reads PDFs, outlines, bullet points, or existing PPTX files as evidence (never edits the input in place)
- Produces a deck plan conforming to a JSON schema: one action title, one bounded claim, evidence, and source references per slide
- Builds new editable 16:9 PPTX files with PptxGenJS 4.0.1, snapshotting every revision
- Runs deterministic QA (qa_layout.py, check_pptx.py) to block template violations
- Renders slide PNGs (native PowerPoint first, LibreOffice/Poppler fallback) for visual review
- Atomically copies the reviewed revision to the output path once both gates pass
- A grad student converting a paper PDF into a 12-slide Chinese lab-meeting deck
- A PhD candidate preparing defense slides with a consistent navy template and red key-finding box
- A lab member rebuilding a journal-club presentation from bullet points while keeping a team-wide visual style
- Extracting content from an existing PPTX and rebuilding a new version rather than editing in place
- Teams where members on Codex or Claude Code need output that looks like the same template
- Users needing native-PowerPoint-fidelity rendering on WSL/Linux — only LibreOffice fallback is supported there
- CI or unattended servers — PowerPoint automation requires a signed-in interactive desktop session
- Lightweight users who want a quick deck without installing the full Python/Node/LibreOffice toolchain
How do you install this skill?
- This assessment is a static source review only; nothing was executed and confidence is low.
- Full functionality requires local PowerPoint, LibreOffice, Node.js and other dependencies; servers and headless environments only support generation and preflight.
- Baidu OCR uploads user document images to a third-party network service; users with sensitive material should enable it cautiously despite the grant/revoke mechanism.
- Publisher identity is unverified; maintenance and updates depend on a personal repository, creating long-term uncertainty.
- Visual review quality depends on host vision or human inspection; a text-only model cannot complete that gate.
- Shell / CLI
- Local filesystem
Python 3.12Node.jspptxgenjs 4.0.1LibreOfficePopplerPowerPoint desktop (optional, for native rendering)
Codex (macOS/Linux) — clone into the skills directory and set up a virtual environment:
mkdir -p ~/.codex/skills
git clone https://github.com/jiadizhunine/deepPPT.git ~/.codex/skills/deep-ppt
cd ~/.codex/skills/deep-ppt
python3.12 -m venv .venv
.venv/bin/python -m pip install -r requirements.txt
npm install -g [email protected]Claude Code (macOS/Linux) — install into ~/.claude/skills:
mkdir -p ~/.claude/skills
git clone https://github.com/jiadizhunine/deepPPT.git ~/.claude/skills/deep-ppt
cd ~/.claude/skills/deep-ppt
python3.12 -m venv .venv
.venv/bin/python -m pip install -r requirements.txt
npm install -g [email protected]Codex (Windows; Claude Code users change .codex to .claude):
git clone https://github.com/jiadizhunine/deepPPT.git "$HOME\.codex\skills\deep-ppt"
py -3.12 -m venv .venv
.venv\Scripts\python -m pip install -r requirements.txt
npm install -g [email protected]How do you use this skill?
Once installed, send your agent any of these to trigger it:
- Use $deep-ppt to create a 12-slide Chinese lab meeting deck from this paper PDF. Keep the deepPPT visual style, include rendered QA, and provide the final PPTX.
- Turn this outline into a 15-slide English conference deck with rendered visual review using deepPPT.
- Extract the content from this existing PPTX and rebuild it as a new defense deck, using the red key-finding box on the key conclusions.
Invoke $deep-ppt in a Skills-capable agent with your topic, source material, and target scenario. Core flow: init creates a run (collecting sources, output path, slide count) → inspect → plan (submit a schema-conforming deck plan) → build with PptxGenJS → check (deterministic QA) → render (slide PNGs) → review (visual review JSON) → complete (delivery). Note: intro-page concept images are disabled by default (disabled_not_requested) and only generated on Codex hosts with ImageGen when explicitly requested; Baidu OCR is a network provider that requires per-task explicit user authorization and auto-revokes when the run ends.
What are this skill's strengths and limitations?
- Deterministic QA scripts block repeated template rules, red boxes on agenda pages, and red-box overuse
- Run state is persisted and resumable via resume after interruption
- Dual delivery gates (deterministic checks plus rendered visual review), with reports bound to the exact PPTX hash
- Supports Codex, Claude Code, and OpenClaw with a unified visual grammar
- Requires a heavier local toolchain: Python 3.12, Node.js, pptxgenjs, LibreOffice/Poppler
- Native PowerPoint-fidelity rendering only on macOS/Windows interactive desktop sessions
- README badge shows v0.1.3 while SKILL.md says v0.2.0 — version info is inconsistent
- Text-only models cannot perform the visual-review step; host vision or human review is required
How does this skill compare with similar options?
Side by side with related skills; every score comes from the same FSRS standard.
| Skill | FS score | Stars | Last updated | License |
|---|---|---|---|---|
| deepPPT this page | 67 · Recommended | ★ 28 | 2mo ago | MIT |
| PPTX Presentation Skill (Open Cowork) | 69 · Recommended | ★ 2.2k | 12d ago | MIT |
| PPTX Presentation Skill ✓ Anthropic · Official | 54 · Use with care | ★ 181k | 16d ago | — |
| PPTX Presentation Skill | 47 · Use with care | ★ 3.1k | 8mo ago | MIT |
| DeepTutor pptx Skill | 58 · Recommended | ★ 41k | 17d ago | Apache-2.0 |
The README names no direct competitors. Compared with letting an LLM improvise a deck, deepPPT's differentiator is codifying layout rules into scriptable deterministic QA and rendered-preview gates.
How did FollowSkills review this skill?
Evidence shows strong least-privilege design: Baidu OCR is off by default, requires task-scoped grant/revocation, auto-revokes on completion; ImageGen availability is decoupled from user intent; credentials are never persisted; output cannot overwrite sources; fingerprint binding prevents reusing stale reports. Deducted for unverified publisher identity, the fact that OCR still uploads user documents to a third party despite the consent flow, and confirmation relying on CLI discipline where no approval UI exists.
Real three-platform CI plus extensive committed unit tests covering lifecycle, permissions, idempotency and fingerprint checks constitute auditable execution evidence, exceeding the static cap of 10. Deducted because nothing was executed in this review, test files are partially truncated, failure-feedback quality on abnormal input is only partially inferable, and visual review depends on host vision or humans, so determinism cannot be fully verified.
Target scenarios (lab meetings, journal clubs, defenses, bilingual academic decks) and trigger conditions are explicit; platform routing boundaries (no native-fidelity claims on WSL/Linux) and non-fit ranges are disclosed; Chinese examples are complete; Baidu OCR is reachable from mainland China. Deducted for heavy multi-dependency local installation (PowerPoint/LibreOffice/Node), limited server/CI usability, and no usage evidence for trigger precision.
Well-layered docs (SKILL.md, bilingual READMEs, style guide, patterns, schemas, examples), version 0.2.0, MIT license, pyproject packaging with CI asserting SKILL.md ships in the wheel. Deducted for no standalone changelog, anonymous 'contributors' authorship, unverifiable publisher identity, and update path dependent on a personal repository.
Atomic delivery of an editable PPTX only after dual gates (deterministic QA + rendered visual review) is a clear value claim, and CI actually rebuilds and validates example PPTX files, surpassing prompt-only alternatives. Deducted because real-task output quality is unverified in a static review, visual review depends on host/human vision, and heavy local dependencies hurt cost/benefit for lightweight users.
Multiple auditable evidence types exist: CI workflow, committed test suite, downloadable example PPTX files, fingerprint-bound reports. Deducted because independent multi-source reproduction is absent, test evidence is only partially visible (truncated files), and no third-party verification exists, capping at 5.
Open a dimension to read why it scored that way
Evidence confidence：Low — Mostly static review, author material or a limited demo; useful for discovery, not high-risk decisions.
See the full review method →