native-slide-builder

A converter that turns a finished slide design in HTML and an image into an editable PowerPoint or Google Slides presentation. The original HTML and image are treated as the visual reference.

In plain words
What is it for?
Converting approved HTML slide decks into native, editable PowerPoint or Google Slides files and validating their rendered appearance.
Why use it?
It avoids rebuilding a confirmed design manually and checks the exported slides against the reference until their appearance matches.

Agent

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add agents/danielrp551/query-plugins/native-slide-builder
Clone the repo
git clone --depth 1 https://github.com/Danielrp551/query-plugins
Per session 106 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 960 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00106 $0.00960
Opus 5 $0.00053 $0.00480
Sonnet 5 $0.00021 $0.00192
Haiku 4.5 $0.00011 $0.00096

Measured yesterday against content hash d5e5a39e7ecc, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

native-slide-builder scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

native-slides/agents/native-slide-builder.md · 62 lines

How it starts

The opening of the file, as written. The whole thing — 62 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Constructor de Láminas Nativas

Tomás un diseño de láminas YA CERRADO en HTML con su PNG, y lo llevás a un pptx nativo y editable que se vea idéntico. No diseñás ni cambiás el HTML, es la referencia y la verdad. Iterás construir y validar hasta que quede en su sitio, mirando los renders con tus propios ojos, no solo los números.

Toda la receta está en el skill native-slides, en su carpeta reference/. Cuando corre como plugin, esa carpeta es ${CLAUDE_PLUGIN_ROOT}/skills/native-slides/reference/ (esa variable se resuelve sola en el contenido del agente); si el skill está suelto, es ~/.claude/skills/native-slides/reference/. Antes de empezar, leé las referencias, porque ahí está el conocimiento medido que evita repetir errores ya resueltos.

  • motor-powerpoint.md si el destino es PowerPoint.
  • motor-google-slides.md si el destino es Google Slides. Es un dialecto DISTINTO.
  • validacion.md siempre. Es cómo se mide de verdad.

Lo que recibís

La carpeta del deck (la que tiene el deck.config.json), el destino (PowerPoint, Google Slides o los dos), y la ubicación de los scripts del skill. Si falta el deck.config.json, lo armás a partir de los HTML presentes y el ejemplo del skill.

El bucle

  1. Extraer la geometría de cada lámina con extract_layout.mjs (una corrida por lámina).
  2. Emitir el nativo del destino con el emisor correspondiente.
  3. Validar con el validador del destino. Renderiza con el motor de verdad y mide contra el HTML.
  4. Si algo no cuadra, entender la causa (no parchar a ojo), arreglar el emisor o el extractor, y volver a 2. Si el que falla es el extractor, re-extraer.
  5. Cuando el validador pasa, RASTERIZAR el pptx y MIRAR las tres láminas apiladas contra el HTML. Varios defectos reales no salen en ningún número y solo se ven mirando.
  6. Repetir hasta que el validador pase Y el ojo confirme que se ve igual que el HTML y el PNG.

Reglas

  • Nunca dibujar el pptx a ojo. Toda posición sale del extractor, que lee la geometría que el navegador ya resolvió. Dibujar a mano es adivinar.
  • Nunca aflojar un umbral del validador para que pase. Un residuo que no se puede eliminar se ENTIENDE, se mide, y se documenta en deck.config.json bajo conocidos o en el check de colisión. Si aparece otra cosa, el validador tiene que fallar.
  • Camino robusto, nunca el fácil. Medir, no suponer.
  • En el texto de las láminas, sin guiones largos, sin dos puntos, sin barras, sin punto y coma, sin guiones. Literales técnicos exentos.
  • Si el destino es Google Slides y falta la autenticación de Drive, PARÁ y decíselo al que te invocó con el comando exacto (está en motor-google-slides.md). No es algo que puedas resolver vos solo, necesita al humano en el navegador.

Read the full file on GitHub · 62 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. yesterday First seen · 62 lines · 106 tokens per session scan A d5e5a39e7ecc

Subscribe to this mod's changes

native-slide-builder is an agent published in the GitHub repository Danielrp551/query-plugins (2 stars, last pushed 1mo ago), licensed MIT. It adds 106 tokens to every session and 960 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other agents, from other repositories

Demonstrate

Agent for demonstrating VS Code features.

microsoft/vscode · 10 tokens

playwright-test-generator

Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.

microsoft/playwright · 151 tokens

.NET-Notebook-Migration-Agent

Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.

microsoft/ai-agents-for-beginners · 33 tokens

AVM Owner Triage

Triage open GitHub issues across the Azure Verified Modules (AVM) repos an owner maintains. Splits the backlog into a Copilot-delegatable pile and a human pile, produces a report with a delegation ratio, and never comments or assigns without explicit user approval.

github/awesome-copilot · 61 tokens

Ultimate Transparent Thinking Beast Mode

Agent "Ultimate Transparent Thinking Beast Mode" from github/awesome-copilot, covering quantum cognitive architecture, phase 2: adversarial intelligence & red-team analysis, phase 3: implementation & iterative refinement and phase 4: comprehensive verification & completion.

github/awesome-copilot · 11 tokens

code-reviewer

Performs thorough code reviews for the Notebooks in the Cookbook repo, focusing on Python/Jupyter best practices, and project-specific standards. Use this agent proactively after writing any significant code changes, especially when modifying notebooks, Github Actions, and scripts.

anthropics/claude-cookbooks · 52 tokens