Media & Document Processing

Complete reference of media processing and document generation capabilities.

Document Parsing

Supported Input Formats

Format	Library	Features
PDF	pdf-parse	Text extraction, multi-page
DOCX/Word	mammoth	Rich text, formatting
RTF	Native	Control code stripping
CSV	Native	Headers, row parsing
XLSX/Excel	Native	Read-only, cell extraction
JSON	Native	Structured data
YAML	Native	Configuration files
XML	Native	Structured markup
HTML	Native	Web content
Markdown	Native	Rich text
Plain Text	UTF-8	All text files

Code File Parsing

All major programming languages with syntax detection:

TypeScript, JavaScript, Python, Java, Go, Rust, C, C++, C#,
Ruby, PHP, Swift, Kotlin, Scala, Shell, SQL, HTML, CSS,
JSON, YAML, XML, Markdown, Dockerfile, Makefile

---

Document Generation

PDF Generation

Libraries: pdf-lib, Playwright

Feature	Status
Multi-page documents	Supported
Custom fonts	Supported (Helvetica default)
Text wrapping	Supported
Headings and body styles	Supported
Color customization (RGB)	Supported
Page sizes (A4, Letter, Legal)	Supported
Landscape/Portrait	Supported
Headers/Footers	Supported
Print backgrounds	Supported

Example:

// Generate PDF from HTML sections
const pdf = await pdfGenerator.generate({
  sections: [
    { type: "heading", content: "Report Title" },
    { type: "body", content: "Report content..." }
  ]
});

Word Documents (DOCX)

Library: docx

Feature	Status
Heading levels	Supported
Text runs	Supported
Paragraphs	Supported
Metadata	Supported

Screenshots (PNG)

Library: Playwright

Feature	Status
Full page capture	Supported
Element screenshots	Supported
Quality settings	Supported
Viewport configuration	Supported

Other Formats

Format	Status	Use Case
SVG	Supported	Vector diagrams
CSV	Supported	Spreadsheet data
JSON	Supported	Structured data
Markdown	Supported	Rich text
Plain Text	Supported	Simple output

---

Planned Document Formats

High Priority

Format	Status	Use Case
PPTX/PowerPoint	Planned	Presentations
XLSX/Excel (write)	Planned	Spreadsheets
LaTeX	Planned	Academic papers

Medium Priority

Format	Status	Use Case
ODT (OpenDocument)	Planned	LibreOffice docs
ODS (OpenDocument)	Planned	LibreOffice sheets
EPUB	Planned	E-books
RTF (write)	Planned	Rich text

---

OCR (Optical Character Recognition)

Active Providers

Provider	Status	Features
Tesseract.js	Supported	LSTM neural network, multi-language

OCR Features

Feature	Status
Block classification	Supported
Confidence scoring	Supported
Bounding boxes	Supported
Word positioning	Supported
Multi-image batch	Supported

Planned OCR Providers

Provider	Status	Notes
Azure Form Recognizer	Planned	Document analysis
Google Vision API	Planned	High accuracy
AWS Textract	Planned	AWS ecosystem

---

Image Processing

Image Formats

Format	Read	Write
PNG	Yes	Yes
JPEG/JPG	Yes	Yes
WebP	Yes	Yes
GIF	Yes	Yes
BMP	Yes	Yes
TIFF	Yes	No
SVG	Yes	Yes

Image Operations

Operation	Tool	Description
Resize	media.transform	Change dimensions
Crop	media.transform	Extract region
Format conversion	media.transform	PNG ↔ JPEG ↔ WebP
Enhance	media.transform	Improve quality
Redact/blur	media.transform	Hide sensitive content

Image Annotations

Type	Status
Highlight areas	Supported
Arrows/pointers	Supported
Text labels	Supported
Rectangles	Supported
Circles/ellipses	Supported
Freehand drawing	Supported
Blur/redact	Supported
Pin markers	Supported

---

Image Generation

Active Providers

Provider	Status	Output
DALL-E (OpenAI)	Supported	PNG 1024x1024

Planned Providers

Provider	Status	Notes
Stable Diffusion	Planned	Open source
Flux	Planned	High quality
Midjourney	Planned	Artistic style

Generation Tools

Tool	Description
media.create	Generate from text prompt
media.transform	Modify existing image
media.analyze	Analyze image content

---

Video Processing

Video Formats

Format	Status
MP4	Supported
WebM	Supported
OGG	Supported
MOV/QuickTime	Supported
AVI	Supported
MKV/Matroska	Supported

Video Features

Library: FFmpeg/FFprobe

Feature	Status
Frame extraction	Supported
Metadata extraction	Supported
Duration detection	Supported
Codec information	Supported
Resolution detection	Supported
FPS detection	Supported
Key frame extraction	Supported
Thumbnails	Supported
Transcoding	Planned

Planned Video Generation

Provider	Status
Runway	Planned
Pika	Planned
Sora (OpenAI)	Planned

---

Audio Processing

Audio Formats

Format	Status
MP3	Supported
WAV	Supported
OGG	Supported
FLAC	Supported
AAC	Supported
WebM Audio	Supported
M4A	Supported

Audio Features

Feature	Status
Metadata extraction	Supported
Duration detection	Supported
Transcription	Planned (requires speech-to-text)

---

Diagrams & Visualization

Diagram Formats

Format	Status	Use Case
Mermaid	Supported	Flowcharts, sequence, state, class
PlantUML	Supported	UML diagrams
Excalidraw	Supported	Hand-drawn style
Graphviz (DOT)	Supported	Graph visualization

Diagram Features

Feature	Status
Theme support	Supported (default, dark, forest, neutral)
Editable diagrams	Supported
SVG export	Supported
Source preservation	Supported

Chart Types

Type	Status
Line charts	Supported
Bar charts	Supported
Pie charts	Supported
Scatter plots	Supported
Area charts	Planned
Heatmaps	Planned

---

Dynamic Card System

Content Types (21)

Type	Description
document	PDF, DOCX, text documents
image	PNG, JPG, SVG with annotations
video	MP4, WebM with frames
audio	MP3, WAV recordings
code	Syntax-highlighted source
markdown	Rich text documents
html	Web content
json	Structured data viewer
csv	Spreadsheet data
form	Dynamic forms
diagram	Mermaid, PlantUML, etc.
blueprint	Design blueprints
spreadsheet	Excel-like data
terminal	CLI output
diff	Code diffs
database	Query results
search	Search results
agent	Agent configs
error	Error displays
link	URL previews
preview	Web previews

Export Formats

Format	Description
PDF	Print-ready documents
DOCX	Microsoft Word
PNG	Screenshot/image
SVG	Vector graphics
CSV	Spreadsheet data
JSON	Structured data
Markdown	Rich text
Plain Text	Simple text

---

Cloud Storage Integration

Supported Providers

Provider	Status
Google Drive	Supported
OneDrive	Supported
Dropbox	Supported
AWS S3	Supported

Features

Feature	Status
Upload	Supported
Download	Supported
Versioning	Supported
Sharing	Supported

---

Document Analysis

Layout Analysis

Feature	Status
Text block detection	Supported
Image region detection	Supported
Table detection	Supported
Form field detection	Planned

Document Tools

Tool	Description
document.read	Extract structured data
document.query	Query specific fields
document.compute	Calculations on data
document.create	Create new document
document.report	Generate report
document.blueprint	Create architecture docs
document.artifact	Create code artifacts
document.spec	Create specifications