Menu
Export research, reports, and analysis to PDF, DOCX, PPTX, CSV, and XLSX with professional formatting, charts, and theming.
Instalar com Codex ou Claude Copie este prompt, cole no Codex, Claude ou outro assistente e deixe que ele revise a página da skill e instale para você.
Baseado na classificação ocupacional SOC
Trust-aware dialogue generation for AI companion relationships
Generate, edit, upscale, variate, and style-transfer images using the AgentOS multi-provider image pipeline with automatic fallback and character consistency.
Difficulty scaling, CR calculations, action economy, adaptive encounter design
Fair adjudication of game encounters — combat, puzzles, social challenges
Renders game state changes into dramatic, immersive prose narrative
Branching quest DAGs, prospective memory triggers, workflow composition
| name | document-export |
| version | 1.0.0 |
| description | Export research, reports, and analysis to PDF, DOCX, PPTX, CSV, and XLSX with professional formatting, charts, and theming. |
| author | Wunderland |
| namespace | wunderland |
| category | productivity |
| tags | ["document","export","pdf","docx","pptx","csv","xlsx","report","slides","charts","spreadsheet","presentation"] |
| requires_secrets | [] |
| requires_tools | ["document_export","document_suggest"] |
| metadata | {"agentos":{"emoji":"📄","homepage":"https://wunderland.sh"}} |
You can export your responses, research, and analysis to professional documents in five formats: PDF, DOCX, PPTX, CSV, and XLSX.
document_exportGenerate a document from structured content. Accepts a format, structured content with sections, and optional configuration.
Formats: pdf, docx, pptx, csv, xlsx
Content structure:
title (required) -- document title for cover pages and metadatasubtitle -- shown below the title on cover pagesauthor -- embedded in document metadatadate -- ISO 8601 date stringtheme -- visual preset: dark, light, corporate, creative, minimalsections[] -- ordered content sections, each containing any combination of:
heading + level (1/2/3)paragraphs[] -- body text with inline markdown (**bold**, *italic*, [link](url))table -- { headers: string[], rows: string[][] }chart -- { type: bar|line|pie|doughnut|area|scatter, data: [...], title? }list -- { items: string[], ordered?: boolean }keyValues -- [{ key, value }] rendered as definition tablesimage -- { url?, base64?, caption?, width? }speakerNotes -- PPTX only, attached as slide noteslayout -- PPTX hint: title, content, two-column, image-left, image-right, chart-full, comparisonOptions: filename, pageSize (letter/a4/legal), orientation (portrait/landscape), coverPage, pageNumbers
Output: { filePath, downloadUrl, previewUrl, format, sizeBytes, filename }
document_suggestCheck whether a response should offer document export. Pure heuristic, no LLM call.
Input: responseText, wordCount, hasTableData, hasSections, isAnalytical
Rules:
Output: { shouldOffer, suggestedFormats[], offerText }
After generating substantive responses, use document_suggest to check if you should offer export:
When the user requests an export (or accepts your offer):
document_export with the structured contentBest for long-form research, analysis, and formal reports.
Same content capabilities as PDF but editable.
Each section maps to one slide. Best for presentations and visual summaries.
speakerNoteslayout hints: title for opener, content for standard, chart-full for data slidesThemes:
corporate -- business presentations, quarterly reviews (navy/grey, Arial)dark -- conference talks, tech demos (navy background, cyan accent)light -- general purpose, academic (white background, blue accent) [default]creative -- marketing, workshops (warm amber, gold accent, Georgia titles)minimal -- data-focused, research (monochrome, generous whitespace)Tabular data only. Each table section becomes a block of rows; multiple tables separated by blank rows.
Multi-sheet Excel workbooks with formatting.
document_suggest detects long analytical response -> shouldOffer: truedocument_export with format: "pdf", structured sections, charts for datadocument_suggest detects tabular data -> suggests CSV, XLSXdocument_export with tables as sectionslayout: "title" for opener, "content" for body, "chart-full" for datatheme: "corporate" for business contextsspeakerNotes with talking pointsdocument_export with format: "pptx"If the user says "export this" without specifying format, offer the most relevant options: