Google AI Studio | Chat/Conversation Markdown-Export (API Based)

Exports the full conversation to Markdown. Uses a recursive scanner to find the chat history regardless of API changes.

当前为 2025-11-29 提交的版本,查看 最新版本

您需要先安装一个扩展,例如 篡改猴Greasemonkey暴力猴,之后才能安装此脚本。

You will need to install an extension such as Tampermonkey to install this script.

您需要先安装一个扩展,例如 篡改猴暴力猴,之后才能安装此脚本。

您需要先安装一个扩展,例如 篡改猴Userscripts ,之后才能安装此脚本。

您需要先安装一款用户脚本管理器扩展,例如 Tampermonkey,才能安装此脚本。

您需要先安装用户脚本管理器扩展后才能安装此脚本。

(我已经安装了用户脚本管理器,让我安装!)

您需要先安装一款用户样式管理器扩展,比如 Stylus,才能安装此样式。

您需要先安装一款用户样式管理器扩展,比如 Stylus,才能安装此样式。

您需要先安装一款用户样式管理器扩展,比如 Stylus,才能安装此样式。

您需要先安装一款用户样式管理器扩展后才能安装此样式。

您需要先安装一款用户样式管理器扩展后才能安装此样式。

您需要先安装一款用户样式管理器扩展后才能安装此样式。

(我已经安装了用户样式管理器,让我安装!)

作者
piknockyou
评分
0 0 0
版本
1.0
创建于
2025-11-29
更新于
2025-11-29
大小
10.1 KB
许可证
暂无
适用于

Google AI Studio | Chat/Conversation Markdown-Export (API Based)

Instantly export complete Google AI Studio conversations to Markdown by intercepting the internal API. Features a native toolbar integration, smart recursive parsing, and zero-scrolling capture.

Features

  • API-Level Extraction: Intercepts the raw ResolveDriveResource network request to capture the exact chat data sent from Google's servers.
  • Instant Capture: No scrolling, no waiting, and no rendering delays. The data is captured the moment the page loads.
  • Smart Recursive Parsing: Uses an advanced algorithm to locate chat history within Google's complex JSON structure, making it resilient to API changes.
  • Native Toolbar Integration: Adds a seamless "Download" button to the existing AI Studio toolbar using Material Design styling.
  • Complete History: Captures every user prompt and AI response in the conversation, ensuring nothing is missed due to UI virtualization.
  • Clean Markdown Output: Generates a formatted .md file ready for archiving or use with other LLM agents.
  • Non-Intrusive: Runs silently in the background and only modifies the DOM to add the single download button.

Installation

Requirements

Steps

  1. Install one of the userscript managers listed above.
  2. Download/Install this script.
  3. The script will automatically activate on https://aistudio.google.com/*.

Usage

Important: Because this script captures data as it loads, you must be on the page when the data fetch happens.

  1. Navigate to any conversation in Google AI Studio.
  2. Refresh the Page (F5).
    • Why? AI Studio fetches the full chat history only once when the page first opens. The script needs to be running to "catch" this file.
  3. Look for the Download Icon in the top-right toolbar (next to the "Run" or "More" buttons).
  4. Click the button to instantly download the Markdown (.md) file.

How It Works

Unlike traditional scrapers that try to read text from the screen, this script uses a more robust approach:

  1. Network Interception: It hooks into the browser's XMLHttpRequest to listen for a specific Google Drive resource request (ResolveDriveResource).
  2. Data Capture: When Google sends the chat history to your browser, the script saves a copy of this raw JSON data.
  3. Recursive Search: It scans the deeply nested JSON structure to find the specific array containing "user" and "model" dialogue, regardless of where Google hides it.
  4. Formatting: It converts the raw data into a clean Markdown format.
  5. Download: It generates a file named after the prompt title (or a timestamp) and triggers a browser download.

FAQ

Q: I don't see the button. A: Refresh the page. The button only appears after the script successfully detects and captures the chat data from the network stream.

Q: Does this scrape the screen? A: No. It reads the raw data file directly from the network traffic. This is faster and more accurate than screen scraping.

Q: Can I filter messages? A: This version captures the entire conversation exactly as it exists in the file to ensure data integrity.

Q: Is it safe? A: Yes. The script runs entirely locally in your browser. It does not send your data to any third-party server; it simply converts the data Google sent you into a text file.