Skip to content
agent-ready-repo

converters

Say “convert this PDF to Markdown” or “turn this Markdown into a branded Word doc” — and the agent handles it. converters moves content between every format your team touches: PDF, Office, HTML, EPUB, email, images, and Mermaid diagrams → Markdown, and Markdown back out to styled HTML, Word, PowerPoint, and Excel.

The file-to-markdown skill runs a no-ML Tier-0 floor (pure-Python parsers for PDF, Office, HTML, EPUB, CSV, ODF, .eml) before reaching for Docling. Every output carries a versioned frontmatter contract — provenance and a quality/confidence signal — so downstream tools know what they received.

The Markdown → Office skills (markdown-to-docx, markdown-to-pptx, markdown-to-xlsx) fill your existing branded templates. Each detects its render library (docxtpl / python-pptx / openpyxl) at runtime and stops with the exact pip install line if it’s absent.

Skills

Section titled “Skills”
  • file-to-markdown — PDF, Office, HTML, EPUB, CSV, ODF, .eml, images → Markdown
  • markdown-to-html — Markdown → styled HTML
  • markdown-to-docx — Markdown → branded Word document
  • markdown-to-pptx — Markdown → branded PowerPoint deck
  • markdown-to-xlsx — Markdown → branded Excel workbook
  • msg-to-markdown — Outlook .msg → Markdown
  • mermaid-renderer — render Mermaid diagrams

Install

Section titled “Install”
agentbundle install --pack converters --scope user <catalogue>

converters is user-scope by default — format conversion is a portable utility, not a project concern.

→ Go deeper: guides/converters/