メニュー
Export, list, and search GitHub Copilot chat sessions from VS Code (both stable and Insiders) to HTML, JSON, JSONL, or text. Use when the user wants to save, share, back up, summarize, or search past Copilot chat conversations.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
Autonomous AI agent workflow — file-based UPD/RESULT cycle
Pull GitHub Copilot usage statistics without UI or screenshots. Two Python CLIs: one reads live credit/quota numbers from GitHub's private Copilot API, the other runs fast, context-free queries over the plugin's per-session `main.jsonl` debug log (tokens, requests, tool calls). Use when the user wants to track AI credits, premium-request usage, token counts, or inspect what a chat session actually did.
Fill a DOCX template interactively — extracts <placeholder> names via CLI, asks the user for each value with vscode_askQuestions, then produces a filled copy. The original template is never modified.
Analyze git commit history to extract SDLC instruction files. Use when: reverse-engineering workflows from commits, extracting coding patterns from diffs, building agent instructions from real code changes, generating automation knowledge base from commit history. Iterates through commits, extracts per-file diffs, invokes Copilot CLI (Claude Opus 4.6) to identify and document reusable SDLC patterns.
Translate any website to any language using Chrome DevTools MCP — the AI model extracts text from the page, translates it, and injects the translation back into the DOM.
Convert a list of markdown files (with images, code blocks, quizzes) into a single landscape DOCX with TOC, page breaks between files, shaded inline-code, and uniform image scaling
| name | copilot-chat-export |
| description | Export, list, and search GitHub Copilot chat sessions from VS Code (both stable and Insiders) to HTML, JSON, JSONL, or text. Use when the user wants to save, share, back up, summarize, or search past Copilot chat conversations. |
Standalone Python tools that read VS Code's internal chat storage and convert
sessions to portable formats. No external dependencies — Python 3 only.
Supports both the legacy .json and the modern .jsonl (delta/CRDT) session
formats.
| Script | Purpose |
|---|---|
scripts/chat_export.py | List workspaces & sessions, export one or many sessions, search across all history |
scripts/export_all.py | Batch-export every session from every workspace, preserving folder structure |
Run from the repository root.
# List all workspaces that have chat history
python ./modules/250-export-chat-session/tools/copilot/scripts/chat_export.py workspaces
# List sessions inside one workspace
python ./modules/250-export-chat-session/tools/copilot/scripts/chat_export.py sessions <workspace_id>
# One session to HTML (default format)
python ./modules/250-export-chat-session/tools/copilot/scripts/chat_export.py export <workspace_id> <session_id> --output-dir ./work/copilot_export --format html
# Other formats: json | jsonl | text
python ./modules/250-export-chat-session/tools/copilot/scripts/chat_export.py export <workspace_id> <session_id> --format json
# All sessions in one workspace
python ./modules/250-export-chat-session/tools/copilot/scripts/chat_export.py export <workspace_id> * --output-dir ./work/copilot_export
python ./modules/250-export-chat-session/tools/copilot/scripts/chat_export.py search "github mcp"
python ./modules/250-export-chat-session/tools/copilot/scripts/export_all.py
python ./modules/250-export-chat-session/tools/copilot/scripts/export_all.py --output-dir ./work/my_backup
python ./modules/250-export-chat-session/tools/copilot/scripts/export_all.py --format json
| Option | Meaning |
|---|---|
--vscode-path PATH | Override the VS Code settings path (auto-detects Code and Code Insiders by default) |
--output-dir DIR | Output directory (default ./work/copilot_export) |
--format FORMAT | html (default), json, jsonl, or text |
Exported sessions may contain secrets, tokens, file contents, or customer data. Treat them like server logs: review before sharing externally, and redact sensitive patterns when exporting for a public audience.