Content Processing Commands

Commands for extracting, transforming, validating, and generating content from documents, web pages, images, and data files.

Content Extraction

Command	Input	Description
extract-pdf	PDF files	Extract PDF to markdown with OCR and template support
extract-web	URLs	Fetch web pages via Playwright to markdown/CSV
extract-youtube-auto	YouTube URL	Extract metadata, description, and transcript

Media Processing

Command	Input	Description
process-image	Images	Resize, crop, optimize for social media platforms
generate-logo-package	PNG logo	Generate multi-platform logo delivery package
images-to-pdf	Images	Convert images to a single PDF

Data Transformation

Command	Input	Description
csv-map	CSV	Map columns using YAML profile
md2mail	Markdown	Convert to Gmail-compatible rich text
generate-presentation	Markdown	Generate branded PPTX from structured markdown

Quality & Validation

Command	Input	Description
validate	Markdown files	Validate naming, frontmatter, and formatting rules
health-check	Repository	Check project health, dependencies, and environment

Anonymization (CATE)

Command	Input	Description
cate transform	Any file	Anonymize sensitive content (PII)
cate batch	Directory	Batch-process multiple files for anonymization
cate analyze	Any file	Detect sensitive entities without changes
inbox-poll	IMAP inbox	Poll email inbox and dispatch by route rules

Typical Workflow

Document Extraction Pipeline

1. Extract: extract-pdf --source ./invoices/ or extract-web --url https://example.com
2. Validate: validate --path ./extracted/ to check formatting compliance
3. Anonymize (if needed): cate analyze report.md then cate transform report.md
4. Distribute: md2mail -i report.md to copy to clipboard for email

Logo Delivery

1. Generate package: generate-logo-package --source logo.png --client acme -o ./output
2. Review: Check manifest.json for items flagged as review
3. Custom sizes: process-image --source logo.png --width 800 --height 600 for one-off sizes