Hi,
I'm a long-time watchlist app user (recently switched to Moviebase) and wanted to suggest a feature that I think would set the app apart: exposing an MCP (Model Context Protocol) server alongside the existing integrations.
Context: MCP is an open protocol (originally from Anthropic, now adopted more broadly) that lets AI assistants call a defined set of tools against an app's API - e.g. list_watchlist, add_to_list, update_rating, sync_lists - instead of requiring users to manually export/import data or scrape web pages. Several productivity and media tools (Notion, Todoist, Spotify) already ship MCP servers for this reason.
For Moviebase specifically, this would solve a real pain point: right now, syncing or reconciling lists across services (e.g. a JustWatch shared list vs. a Moviebase list) requires manual copy-pasting, since there's no clean API/export path an AI assistant can reliably parse. A thin MCP wrapper around your existing backend (even read-only to start: list/search/get-details) would let users manage their watchlists conversationally through any MCP-compatible AI assistant.
Given Trakt's existing API, I imagine most of the groundwork (auth, data model) may already exist - the main lift would be the MCP tool definitions and a hosted endpoint.
Happy to be a test user if this is ever on your roadmap. Thanks for building Moviebase - it's a great piece of software.
Best,
Lech
