A local, low-cost extraction tool for RAG and AI agents
Vidilearn is an tool for preparing source material for RAG systems and AI agents on a local machine. It can extract YouTube , subtitles, chapter information, article text, and structured metadata. It is designed to work without API keys and to fit into automation workflows.
It also includes MCP server support so AI tools can connect to it more directly. A benchmark snapshot reports a RAG hit rate of 94.2%, precision of 92.1%, and an F1 score of 0.931. Claude is still slightly better on raw accuracy, but Vidilearn is presented as getting close while running at near-zero cost.
The tool uses Node.js, , and pipelines, and can be installed with npm i vidilearn.
Key points
- Vidilearn is an , extraction tool.
- It prepares YouTube , subtitles, chapters, articles, and metadata for RAG and s.
- It avoids API keys, which can lower cost at scale.
- MCP server support helps connect it to AI tools and agent systems.
- Reported benchmark numbers are 94.2% RAG hit rate, 92.1% precision, and 0.931 F1 score.