Agent skill

Ppt Visual Replica

by ZhiweiWei-NAMI in ZhiweiWei-NAMI/PPT-Visual-Replica

Create and edit editable PowerPoint decks, recreate slides or infographics from reference images, and develop scientific talks, teaching slides, and thesis defenses by generating slide drafts with…

MITAuto-check passedDocuments & Office

Install Ppt Visual Replica

skills CLI
$ npx skills add ZhiweiWei-NAMI/PPT-Visual-Replica --skill ppt-visual-replica -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install ZhiweiWei-NAMI/PPT-Visual-Replica ppt-visual-replica --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/ZhiweiWei-NAMI/PPT-Visual-Replica.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skill/ppt-visual-replica .claude/skills/ppt-visual-replica && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
ppt-visual-replica
GitHub stars
135
Token cost
~1.8k tokens
SKILL.md length
895 words
Files
16 (incl. scripts, references)
Skills in repo
1
Repo updated
First seen
Licence
MIT

At a glance

Create and edit editable PowerPoint decks, recreate slides or infographics from reference images, and develop scientific talks, teaching slides, and thesis defenses by generating slide drafts with…

  • Tasks that involve PowerPoint presentations
  • SKILL.md covers Choose the task, Scope and source material, Required order for new… and Choose the simplest suitable…, plus 3 more sections
  • Runs Python scripts from its folder
  • Tasks that involve Slides and decks

What it does

Ppt Visual Replica is an agent skill from ZhiweiWei-NAMI/PPT-Visual-Replica. Create and edit editable PowerPoint decks, recreate slides or infographics from reference images, and develop scientific talks, teaching slides, and thesis defenses by generating slide drafts with imagegen first and then reconstructing editable PPTX. Choose visual fidelity, scientific evidence, and editability according to the user's requested deliverable.

Its SKILL.md is about 1.8k tokens, which your agent loads only when the skill is triggered. The skill folder holds 18 other files, including scripts and reference files (for example `agents/openai.yaml`, `references/asset-review.md` and `references/reference-reconstruction.md`).

It sits in Documents & Office, covering PowerPoint presentations, Slides and decks and Image generation. It works with Microsoft PowerPoint. The repository describes itself as: Strict image-first workflow for recreating user-provided infographic or presentation reference images as editable PowerPoint decks. The licence is MIT.

When your agent uses it

  • Tasks that involve PowerPoint presentations
  • Tasks that involve Slides and decks
  • Tasks that involve Image generation

Example prompts

  • “/ppt-visual-replica”

Requirements

  • Python 3

What it can do on your machine

Read from SKILL.md and the folder at commit 0e3d24d. It shows what the files ask for, not the result of running them.

  • Tool permissions

    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.

  • Runs code

    Ships 10 files in scripts/ (Python), which the agent can run.

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Ppt Visual Replica loads about 1.8k tokens when it runs, and up to ~8.5k if it reads all its reference files. Until then it costs about 94 tokens; SKILL.md has 895 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~94
When it runs · the whole SKILL.md, loaded when a task matches
~1.8k
With references · SKILL.md plus every file in references/, read only if the agent opens them
~8.5k

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.

Safety

Auto-check passed

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.

SKILL.md

The full file from ZhiweiWei-NAMI/PPT-Visual-Replica at commit 0e3d24d, republished under its MIT licence (© ZhiweiWei-NAMI). 895 words, ~1,801 tokens.

Download SKILL.mdSave it as .claude/skills/ppt-visual-replica/SKILL.md (or your agent's skills folder). This skill also uses 15 other files; get the full folder from GitHub.
name
ppt-visual-replica
description
Create and edit editable PowerPoint decks, recreate slides or infographics from reference images, and develop scientific talks, teaching slides, and thesis defenses by generating slide drafts with imagegen first and then reconstructing editable PPTX. Choose visual fidelity, scientific evidence, and editability according to the user's requested deliverable.

PPT Visual Replica

Produce the requested presentation with readable content, faithful visuals where a reference is supplied, and the requested level of editability. This skill combines visual reconstruction with scientific presentation planning.

Choose the task

  • Reference reconstruction: Match the supplied image or deck's hierarchy, proportions, typography, colors, and meaningful objects. See reference reconstruction.
  • Scientific or teaching presentation: Organize the supplied papers, data, notes, and figures; use the imagegen skill to generate a visual draft of every slide first; then reconstruct those drafts as editable PPTX. This order is required. Follow scientific presentations.
  • Existing deck edit: Preserve its theme, master layouts, content order, and unmodified slides unless the requested change requires otherwise.
  • Strict generated-asset reconstruction: Use strict asset workflow only when the user explicitly requests its per-asset generation, residual tracking, and full audit artifacts. It is a specialized compatibility mode, not a prerequisite for ordinary delivery.

These modes can be combined: for example, recreate the visual style of a reference while writing a scientific talk. Load only the relevant references.

Scope and source material

Use the user's format, template, language, slide count, duration, and supplied sources. Infer routine choices from the material; ask only when missing information materially changes the result. A PPTX request means an editable deck, not a PDF assembled from slide images.

For new decks without a specified size, 16:9 is a reasonable starting point. Respect an existing deck's dimensions. Use installed fonts that support the content language, and report substitutions that affect fidelity.

Required order for new scientific and teaching decks

Source material → slide content plan → imagegen full-slide visual drafts → editable reconstruction → rendered comparison and source verification.

Every planned slide needs a generated draft before reconstruction. Do not substitute a cover image, a few generated icons, or direct PowerPoint coding for this stage. Save the selected drafts with the project. Generated figures and text must not override real data, equations, or citations. Native shapes, vectors, and supplied assets remain valid choices during reconstruction. This requirement applies to new decks from source material; existing-image reconstruction and targeted edits retain their own routes.

Choose the simplest suitable representation

  • Keep titles, labels, equations where supported, tables, and explanatory text editable.
  • Use native shapes, grouped objects, and connectors for diagrams and simple icons when they reproduce the intended appearance.
  • Build charts from real data when it is available. Never manufacture measurements to imitate a chart.
  • Reuse supplied figures and suitable licensed assets. Preserve sources and attribution where required.
  • Use vector artwork when the target application supports it reliably. If conversion to a bitmap is necessary, disclose the resulting editability limit.
  • For new scientific and teaching decks, use the installed imagegen skill for full-slide drafts before reconstruction. Additional isolated illustrations are optional. Follow imagegen's built-in-tool and fallback rules; do not silently skip the required generation stage.
  • Treat an imported picture as an editable picture object, not editable internal geometry. State this distinction when it matters.
  • Do not flatten a whole slide into an image when the user expects editable text and components. A supplied photograph or complex scientific figure may remain a single image if that fits the request.

Use uniform scaling for images. Keep process arrows as connectors where practical. Group objects according to how the user will edit them, not according to a universal one-asset-per-symbol rule.

Show full SKILL.md (349 more words)Show less

Build and review

Choose an available PowerPoint library, application automation, or presentation tool that supports the requested features. The bundled builder is a single-slide strict-mode helper; it is not a general multi-slide presentation engine.

Use visual hierarchy, spacing, contrast, and legible labels to support the message. Avoid fixed quotas for images, bullets, colors, or slides. A text-only slide can be appropriate.

Render the completed deck with an available renderer and inspect each created or modified slide for clipping, overlap, font substitution, unreadable equations, and distorted images. Compare reference reconstructions against the supplied reference. Check native object types when editability matters. After a small correction, recheck affected slides; broaden review if shared layouts or fonts changed.

If rendering or application verification is unavailable, report that limitation and the checks actually performed. Do not describe an unrendered deck as visually verified.

Deliver the requested presentation and useful previews or source assets. Keep manifests, residuals, and diagnostic files only when they support the requested audit or continued editing. Do not make a large evidence package a condition of ordinary delivery.

Optional asset review

For a batch of image assets, use asset review to compare reference crops and candidates, identify missing inventory items, and keep content-bound review records. The agent can review and continue; require user confirmation only when the user requested it or a material ambiguity needs their decision. Generating a wall is not approval.

Bundled tools

Run scripts with --help for their arguments. Paths below are relative to this skill's root.

  • scripts/crop_reference.py: crop regions for inspection or asset generation.
  • scripts/asset_review.py: optional reference/candidate preview wall and persistent review records; see the asset review guide.
  • scripts/grid_cut.py: split a generated asset grid and inspect transparency/borders.
  • scripts/align_from_redboxes.py, scripts/balance_text_lines.py: adjust manifests used by the strict builder.
  • scripts/generate_prompt_pack.py, scripts/subtract_assets.py, scripts/build_pptx.py, scripts/validate_delivery.py: the strict generated-asset pipeline described in its reference. Their narrow schemas and validation rules apply only to that mode.
  • scripts/audit_skill.py: check this skill for hard-coded local paths and unwanted domain residue.

For ordinary native diagrams or multi-slide decks, use an appropriate presentation tool directly; do not force their contents into the strict pipeline's image-only semantic schema.

© ZhiweiWei-NAMI, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 15 other files (scripts, references) in skill/ppt-visual-replica of ZhiweiWei-NAMI/PPT-Visual-Replica.

  • SKILL.md
  • agents/openai.yaml
  • references/asset-review.md
  • references/reference-reconstruction.md
  • references/scientific-presentations.md
  • references/strict-asset-workflow.md
  • scripts/align_from_redboxes.py
  • scripts/asset_review.py
  • scripts/audit_skill.py
  • scripts/balance_text_lines.py
  • scripts/build_pptx.py
  • scripts/crop_reference.py
  • scripts/generate_prompt_pack.py
  • scripts/grid_cut.py
  • scripts/subtract_assets.py
  • scripts/validate_delivery.py

Open the folder on GitHubat commit 0e3d24d

Compare with similar skills

Ppt Visual Replica 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.

Ppt Visual Replica compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Ppt Visual Replica this skillZhiweiWei-NAMI/PPT-Visual-Replica135—~1.8kAutomated safety check: PassMIT
Gpt Image2 PptJuneYaooo/gpt-image2-ppt-skills1.3k—~8.9kAutomated safety check: NotesApache-2.0
Ppt Image2 Editable Rebuildwwe-dog/ppt-image2-editable-rebuild200—~2.2kAutomated safety check: PassUnlicense
EasyslidesRimagination/easyslides138—~12kAutomated safety check: PassMIT
Img2pptxLancelot-Xie/img2pptx163—~5.9kAutomated safety check: PassApache-2.0
PPTX Generatorcoleam00/ai-transformation-workshop147—~9.9kAutomated safety check: PassNone

Similar skills

  • Gpt Image2 Ppt

    JuneYaooo/gpt-image2-ppt-skills

    Generate visually striking PPT slides via OpenAI's gpt-image-2 -- use any style in styles/<collection/STYLEID.md or mimic a user-supplied .pptx template; outputs high-res slide PNGs and a 16:9 .pptx.

    1.3k GitHub stars~8.9k tokensUpdated 1 mo ago
    Documents & OfficeAuto-check: notes
  • Ppt Image2 Editable Rebuild

    wwe-dog/ppt-image2-editable-rebuild

    Create editable PowerPoint decks or pages from image2/imagegen reference PNGs while keeping text and structure editable.

    200 GitHub stars~2.2k tokensUpdated 4 mo ago
    Documents & OfficeAuto-check passed
  • Easyslides

    Rimagination/easyslides

    Use EasySlides to create, distill, edit, review, and export editable academic PPTX presentations, including 蒸馏PPT、提取模板、复用模板和答辩PPT。

    138 GitHub stars~12k tokensUpdated 28 days ago
    Documents & OfficeAuto-check passed
  • Img2pptx

    Lancelot-Xie/img2pptx

    Reconstruct raster reference images as one-slide editable, modular, auditable PPTX files containing a complete SVG.

    163 GitHub stars~5.9k tokensUpdated 1 mo ago
    Documents & OfficeAuto-check passed
  • PPTX Generator

    coleam00/ai-transformation-workshop

    Generate and edit presentation slides as PPTX files. An agent skill from coleam00/ai-transformation-workshop.

    147 GitHub stars~9.9k tokensUpdated 1 mo ago
    Documents & OfficeAuto-check passed
  • Nano Banana

    EvoScientist/EvoSkills

    Generate professional presentation slides and high-quality illustrations using Gemini image generation API (Nano Banana 2), with interactive browser-based review and iterative editing.

    478 GitHub starsUsed in 2 repos~3.5k tokens
    Documents & OfficeAuto-check passed

Questions about Ppt Visual Replica

What does Ppt Visual Replica do?

Create and edit editable PowerPoint decks, recreate slides or infographics from reference images, and develop scientific talks, teaching slides, and thesis defenses by generating slide drafts with…. Ppt Visual Replica is an agent skill from ZhiweiWei-NAMI/PPT-Visual-Replica. Create and edit editable PowerPoint decks, recreate slides or infographics from reference images, and develop scientific talks, teaching slides, and thesis defenses by generating slide drafts with imagegen first and then reconstructing editable PPTX.

When should I use Ppt Visual Replica?

Ppt Visual Replica fits situations like: tasks that involve PowerPoint presentations; tasks that involve Slides and decks; tasks that involve Image generation.

How do I install Ppt Visual Replica in Claude Code?

Run `npx skills add ZhiweiWei-NAMI/PPT-Visual-Replica --skill ppt-visual-replica -a claude-code`. Or copy the skill folder (skill/ppt-visual-replica in ZhiweiWei-NAMI/PPT-Visual-Replica) into .claude/skills/ppt-visual-replica in your project. Claude Code loads it when a task matches its description.

How do I install Ppt Visual Replica in Codex?

Run `npx skills add ZhiweiWei-NAMI/PPT-Visual-Replica --skill ppt-visual-replica -a codex`. Or copy the skill folder (skill/ppt-visual-replica in ZhiweiWei-NAMI/PPT-Visual-Replica) into .agents/skills/ppt-visual-replica in your project. Codex loads it when a task matches its description.

Can I use Ppt Visual Replica in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add ZhiweiWei-NAMI/PPT-Visual-Replica --skill ppt-visual-replica -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/ppt-visual-replica, .gemini/skills/ppt-visual-replica, .github/skills/ppt-visual-replica and .opencode/skills/ppt-visual-replica in your project.

What does Ppt Visual Replica need to run?

Going by SKILL.md and its folder, Ppt Visual Replica needs Python for the scripts in its folder. Our summary lists: Python 3.

Does Ppt Visual Replica access the network?

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.

Is Ppt Visual Replica safe to install?

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.

What licence does Ppt Visual Replica use?

Ppt Visual Replica is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Ppt Visual Replica use?

About 1.8k tokens (SKILL.md is roughly 7.2k 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 6.7k tokens, read only when the agent opens those files.

What are the alternatives to Ppt Visual Replica?

Skills that share tags, products or a category with Ppt Visual Replica: Gpt Image2 Ppt (JuneYaooo/gpt-image2-ppt-skills, 1.3k stars), Ppt Image2 Editable Rebuild (wwe-dog/ppt-image2-editable-rebuild, 200 stars), Easyslides (Rimagination/easyslides, 138 stars) and Img2pptx (Lancelot-Xie/img2pptx, 163 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Ppt Visual Replica?

ZhiweiWei-NAMI (a GitHub user) maintains it in ZhiweiWei-NAMI/PPT-Visual-Replica, which has 135 GitHub stars. The repository was last updated on September 16, 2026.

Source: ZhiweiWei-NAMI/PPT-Visual-Replica on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.