FinditAgent
This Windows tool lets AI agents research large collections of local files without a vector database, working like an enhanced grep that supports almost any file format. The user asks a question, the AI creates the search strategy and queries, and FinditAgent acts as an archivist, executing the searches and sending only results or UTF-8 excerpts back to the AI without ever giving it direct access to the local file system. It offers predefined connections to most cloud model APIs or to locally running AI via Ollama, installs with one click, and is available on GitHub free for private and experimental use. The author is polishing documentation and testing reliability on archives beyond their own.
Shared over time
Each entry reflects what was shared on that date. Details may have changed since then.
FinditAgent
This Windows tool lets AI agents research large collections of local files without a vector database, working like an enhanced grep that supports almost any file format. The user asks a question, the AI creates the search strategy and queries, and FinditAgent acts as an archivist, executing the searches and sending only results or UTF-8 excerpts back to the AI without ever giving it direct access to the local file system. It offers predefined connections to most cloud model APIs or to locally running AI via Ollama, installs with one click, and is available on GitHub free for private and experimental use. The author is polishing documentation and testing reliability on archives beyond their own.
Audience: People who work with large, messy document collections
Posted by @Michael_Houben · Original comment ↗
“final polishing of FinditAgent, a Windows tool which allows AI agents to research large collections of local files without vector database (more like grep on steroids, for almost any file-format) - and without giving the AI any access to local file systems.”