Relentlessly dig through online sources to uncover comprehensive insights on any topic. Research Badger compiles thorough reports by drafting research plans, searching the web, and reading web pages with tenacious attention to detail.
This plugin was forked from TypingMind Deep Research and uses Jina AI for both web search and content extraction. You only need a single Jina AI API key to start using it.
- Open TypingMind and navigate to Plugins
- Click Import plugins button
- Paste this URL into the import field:
https://github.com/wimpysworld/plugin-research-badger - Click Import to add Research Badger to your TypingMind instance
- Download the
plugin.jsonfile from this repository - Open TypingMind → Plugins → Create Plugin
- Select Import plugin from JSON
- Upload the downloaded
plugin.jsonfile
After installation, you'll need to configure the plugin with your API key:
- Open TypingMind → Plugins
- Find Research Badger and click Settings
- Enter your Jina AI API Key: Get your free key at jina.ai/?sui=apikey
- Select your preferred Digging Intensity (see below)
- Save your settings
Note: Jina AI offers a generous free tier suitable for testing and moderate use.
- Snuffle: Research Badger will attempt to extract data from online sources using the extract tool from Jina AI Reader, which saves tokens at a cost of reduced data accuracy. This is the default mode.
- Burrow: Research Badger will read the full content of the online sources to determine the final answer. If the web pages have a lot of content, it can risk consuming a lot of tokens, which is expensive and may exceed the model's context length limit.
- Instinct: Research Badger will run in Snuffle mode by default, but may opt in to read full web page content when absolutely needed to get the most accurate answer.
The Research Badger plugin is customisable. You can duplicate this plugin and add your own tools to enhance the quality of the final report. The system instructions and prompts are also available in the plugin source.
By default, the Research Badger plugin comes with 3 tools: Research Plan, Search Web, and Read Web Page. Depending on your specific needs, you can add more tools like reading from a tweet, searching from a private database, or controlling a browser. The TypingMind plugin system allows you to add multiple tools to the Research Badger plugin and supports various ways to implement your tools using JavaScript, HTTP requests, or MCP.
Learn how to develop TypingMind plugins here.
This plugin may use multiple tools in multiple turns. A deep research task can run for up to 2 minutes. Please monitor the token usage closely to avoid consuming too many tokens. You can also modify the plugin to make it more token-efficient by adding your own tools to perform web searches and read web pages.
Compare the current latest EV cars and their prices available for purchase in the EU.