Flint Note

Agent-first note-taking application that provides complete access to a structured note management system with wikilink support, hybrid search capabilities, and vault organization for building knowledge bases and research workflows.

Agent-first note-taking application with MCP server built by Tim Disney that provides AI assistants with complete access to a structured note management system featuring note types, wikilink support, and hybrid search capabilities. The implementation uses TypeScript with SQLite for storage and includes a vault system for managing multiple note collections, automatic link extraction and bidirectional linking, content hash-based optimistic locking for safe concurrent updates, and comprehensive search tools including regex, metadata filtering, and direct SQL queries. Designed for AI workflows requiring structured knowledge management with features like batch operations, note type schemas with custom metadata fields, cross-vault operations, and MCP resources for read-only access to notes and collections, making it valuable for building AI-powered knowledge bases, research workflows, and collaborative note-taking systems where agents need programmatic access to interconnected notes with rich metadata and linking capabilities.