A zero-dependency file-to-context bridge for Local AI Models.
Built with curiosity and a focus on meaningful tech by Kuldeep Singh.
A seamless, zero-dependency VS Code extension that converts complex documents (PDF, Word, Excel, PowerPoint) into token-optimized Markdown. Designed specifically to bridge the gap between your local file system and local AI models, eliminating the need for paid cloud document parsers.
- Parse for AI (Copy to Clipboard): Right-click any supported file and instantly copy the formatted Markdown to your clipboard, ready to paste into your LLM chat (Continue.dev, LM Studio, etc.).
- Parse for AI (Save as .md): Right-click to extract the Markdown and save it alongside the original file.
- Token Awareness: Automatically warns you if the parsed Markdown exceeds your local model's context window.
- Powered by Microsoft MarkItDown: Uses state-of-the-art document conversion algorithms under the hood, natively bundled as a lightweight executable so you don't need Python installed.
Since this extension is distributed via GitHub Releases, you can install it manually in less than a minute:
- Go to the Releases page.
- Download the latest
markitdown-vscode-1.0.0.vsixfile under Assets. - Open VS Code and navigate to the Extensions view (
Ctrl+Shift+X). - Click the
...(Views and More Actions) menu in the top right corner of the Extensions panel. - Select Install from VSIX...
- Select the
.vsixfile you just downloaded. You're ready to go!
- Open your VS Code File Explorer.
- Right-click a supported file (
.pdf,.docx,.xlsx,.pptx, etc.). - Select "Parse for AI".
This extension contributes the following settings:
markitdown.tokenWarningThreshold: The maximum number of tokens before triggering a context-window warning (default: 8000).
Kuldeep Singh (BSc Computer Science, University of Southampton '29) is a technologist passionate about AI, geopolitics, and digital journalism. Creator of VisionMate (AI accessibility) and WorldFocusIndia.