This is your bridge to LlamaIndex's document parsing service for pulling structured content out of PDFs, Word docs, PowerPoint slides, and spreadsheets. It's a two step process: upload your file, get an ID, then parse with your chosen tier (fast, cost effective, agentic, or agentic plus). The real value shows up in complex documents where you need tables, charts, or mixed layouts extracted cleanly as markdown or JSON. You can pass custom prompts to guide extraction, like translating German invoices or summarizing technical specs. Requires a LlamaCloud API key and handles the TypeScript scaffolding for you, including auth for downloading extracted images.
npx skills add https://github.com/run-llama/llamaparse-agent-skills --skill llamaparse