Skip to content
| Marketplace
Sign in
Visual Studio Code>Formatters>Copilot Chat to MarkdownNew to Visual Studio Code? Get it now.
Copilot Chat to Markdown

Copilot Chat to Markdown

imperium-dev

|
9 installs
| (0) | Free
Export GitHub Copilot chat history (JSON/JSONL) to Markdown. Ideal for Obsidian, Notion, or archiving.
Installation
Launch VS Code Quick Open (Ctrl+P), paste the following command, and press enter.
Copied to clipboard
More Info

Copilot Chat to Markdown

Export your GitHub Copilot chat sessions to clean, usable Markdown.

Recover your AI conversations from VS Code's internal storage or exported files and turn them into permanent knowledge notes for Obsidian, Logseq, Notion, or your personal docs.

Demo

🚀 Features

  • 🔍 Auto-Discovery: Automatically finds Copilot chat sessions hidden in VS Code's workspaceStorage across Windows, macOS, and Linux.
  • 📂 Universal Support: Robustly handles both legacy (.json) and modern (.jsonl) Copilot storage formats.
  • 🔁 Manual Conversion: Convert any exported Copilot JSON/JSONL file to Markdown.
  • 📦 Bulk Export: Export all chat sessions from your current workspace to a structured .wingman directory with a single command.
  • 📝 Clean Markdown: Converts complex internal JSON data into human-readable Markdown.
    • Preserves code blocks and language syntax highlighting.
    • Clearly distinguishes between User and Copilot messages.
    • Terminal Output: Captures and formats executed terminal commands and their output.
    • Handles streamed and chunked responses accurately.
  • ⚡ Instant Access: Fast "Quick Pick" interface lists all your sessions, sorted by recent activity and grouped by workspace.
  • 🔒 Local & Private: Runs entirely locally. No API keys required, no telemetry, and no data leaves your machine.

💡 Use Cases

  • Personal Knowledge Management (PKM): "I solved this with Copilot last week, but I lost the chat." -> Now you can save it to your Obsidian vault.
  • Documentation: Quickly turn an architectural discussion or code explanation into a DOCS.md file for your repo.
  • Team Sharing: Share a complex debugging session or solution with a colleague without sending multiple screenshots.
  • Audit Trail: Keep a version-controlled history of AI decisions and code generation.
  • Wingman: Create a comprehensive .wingman directory in your project root, containing all your AI chat history and project metadata. This is perfect for archiving, analyzing, or feeding into other AI tools.

📖 How to Use

Exporting from Workspace History

  1. Open the Command Palette (Ctrl+Shift+P / Cmd+Shift+P).
  2. Choose one of the following commands:
    • Export Copilot Chat: Current Workspace: Shows chats only from the currently open project.
    • Export Copilot Chat: All Workspaces: Scans all your local VS Code history and groups chats by workspace.
    • Export Copilot Chat: Backup to .wingman (Bulk Export): Automatically exports all chat sessions from the current workspace into a .wingman folder at the root of your project. This includes a history/ subdirectory with Markdown files and a .project.json file with metadata.
  3. Select the session you want to export (for single export commands).
  4. The chat will open instantly in a new Markdown editor tab.
  5. Review the content and Save (Ctrl+S / Cmd+S) it to your preferred location.

Converting an Existing File

  1. Run Export Copilot Chat: Convert JSON/JSONL File....
  2. Select a .json or .jsonl file from your computer.
  3. The extension will parse it and open the Markdown version in a new tab.

⚙️ Extension Settings

  • copilot-pkm-bridge.defaultExportFolder: (Coming Soon) Define a default folder (e.g., ${workspaceFolder}/docs) for quicker saving.
  • copilot-pkm-bridge.includeMetadata: (Coming Soon) Toggle inclusion of YAML frontmatter (dates, model info) in the exported file.

📦 Requirements

  • Visual Studio Code (v1.90.0 or newer recommended).
  • GitHub Copilot Chat extension installed and used (so that chat history exists locally).

🔧 Known Issues

  • If you have moved or renamed workspaces, some sessions might be listed under "Unknown Workspace" until you open that workspace again.
  • Very large sessions (>50k tokens) might take a second or two to parse.

Enjoy building your AI knowledge base!

  • Contact us
  • Jobs
  • Privacy
  • Manage cookies
  • Terms of use
  • Trademarks
© 2026 Microsoft