Levirge

Shared MCP memory for AI agents: knowledge base, skills, todos and agent-to-agent handoffs.

Web search and page reading for AI agents, including JSON APIs, RSS feeds and YouTube.

Web search and page reading for AI agents. Search the web, open any page or JSON API, read RSS feeds and YouTube transcripts, and get the full text back with a source link for every result. **Search** returns ranked results with titles, links and snippets, ready to cite. **Fetch** opens any URL your agent builds — web pages, JSON REST APIs, RSS and Atom feeds — and returns the full content: pages as clean markdown, JSON and XML as the raw, parseable body. Pages that need JavaScript, refuse pla

Shared long-term memory for AI agents: a searchable knowledge base, reusable skills, todos, reminders and agent-to-agent handoffs, scoped per project vault, over one remote MCP endpoint. **Search before answering, capture what you learn.** Captured facts are deduplicated by content, consolidated into per-entity summaries that cite their sources, and rebuilt when a source changes, so answers stay traceable instead of drifting. **Vaults.** Knowledge is partitioned per project, product or client: