Side-by-side comparison of two Model Context Protocol servers — install paths, tools, quality signals, and directory engagement so you can pick the right one for Claude, Cursor, and other MCP clients.
Persistent, local-first AI memory: a semantic knowledge graph of plain Markdown files that humans and LLMs both read and write. Works with any MCP client, with optional cloud sync and team workspaces.
Inspectable memory substrate for Claude Code. Three read-only MCP tools (searchpairs, compactsession, substrateinfo) over a local-first, signal-scored parse of /.claude/projects/. Per-pair scores are numpy columns on disk, not opaque vectors in a service. Zero outbound calls (CI-enforced).