Search across screen, voice, and clipboard entries to find content semantically related to any query. Returns a unified ranked list with source tags for open-ended recall spanning multiple data types.
Enables non-vision LLMs to understand images, extract text via OCR, and parse documents through a unified MCP interface, with local-first processing and optional OpenAI-compatible channels.
Adds safety hooks to AI coding tools by providing enforced proxy tools (safe_write, safe_edit, safe_bash, safe_read, safe_delete) that validate operations through a configurable hook chain before execution, preventing destructive commands, file deletions, and secret exposure.
Search documents using semantic understanding to find relevant content based on meaning rather than keywords. Understands natural language queries and returns ranked passages with source information.
Create a new document collection to organize related documents and enable semantic search across their contents. Returns a collection ID for managing documents.
Discover keyword ideas for any business or location. Input a seed keyword and geographic area to receive related keywords with search volume and other metrics.