In-depth architectural comparison of the Agentcairn and LINZA MCP MCP servers. Compare execution transports, security boundaries, tool capabilities, quality scores, and ready-to-paste client installation snippets for Claude, Cursor, Windsurf, and VS Code.
At a Glance & Executive Verdict
Agentcairn
Knowledge & Memory · Local stdio
Quality: 49/100 (Fair) | Auth: No auth required
LINZA MCP
Knowledge & Memory · Local stdio
Quality: 59/100 (Good) | Auth: No auth required
Verdict Summary: Choose Agentcairn if you need specialized Knowledge & Memory tools running via a local process. Choose LINZA MCP if your workspace requires Knowledge & Memory integration with local subprocess execution. Both servers can be configured concurrently in your client's mcpServers manifest.
Which MCP Server Should You Choose?
Choose Agentcairn when:
You need dedicated capabilities in the Knowledge & Memory domain.
You prefer local stdio subprocess transport architecture.
Your security boundary fits: No auth required (Free / Open Source).
Primary tools included: Plain Markdown Obsidian vault as canonical memory store, Rebuildable DuckDB index for hybrid BM25, vector, and graph recall, Non-lossy memory capture with supersession and secret redaction.
Local-first agent memory: a plain-Markdown Obsidian vault is the source of truth, with a rebuildable DuckDB index for hybrid BM25 + vector + graph recall. Non-lossy capture with secret redaction; works across Claude Code, Codex, Cursor, and any MCP host. uvx agentcairn
Local-first review-gated sidecar for Markdown workspaces. Indexes notes and incoming artifacts into SQLite, proposes domains, links, memory, and growth cards with evidence, and keeps source files unchanged until explicit review. pip install linza-mcp
Category & Scope
Tools & Capabilities Breakdown
Agentcairn Tools (6)
Plain Markdown Obsidian vault as canonical memory store
Rebuildable DuckDB index for hybrid BM25, vector, and graph recall
Non-lossy memory capture with supersession and secret redaction
Cross-agent shared vault for consistent context
Memory provenance, currency, and importance metadata
Integration with Claude Code, Codex, Cursor, and MCP hosts
LINZA MCP Tools (7)
Ready-to-Paste Client Configurations
Paste either (or both) of these JSON server blocks into your client config file (e.g. claude_desktop_config.json or ~/.cursor/mcp.json).
Agentcairn is categorized under Knowledge & Memory and uses a local stdio subprocess. In contrast, LINZA MCP belongs to Knowledge & Memory using local stdio subprocess. Select Agentcairn when you need capabilities focused on knowledge & memory and LINZA MCP when you require tools for knowledge & memory.
Rebuild the LINZA sidecar index for the configured Markdown vault. Use after first setup or after notes change; writes only .linza/linza.db and does not modify note bodies.
search
Find relevant indexed Markdown notes by semantic/lexical query. Use before read_file when the path is unknown; may record search history in the LINZA sidecar.
read_file
Read one vault-relative Markdown file exactly as stored. Use after search or when the path is known; does not write files or sidecar state.
get_stats
Return quick LINZA sidecar status: indexed file count, profiles, semantic bridges, and active profile. Use for health checks.
scan_vault
Run a read-only vault diagnostic over files, links, metadata, and LINZA setup. Use at first contact or when deciding what to fix next.
guide_next_steps
Read the current LINZA state and recommend the next safe onboarding/review step in plain language. This is the navigator for agents and users; it does not write files.
agent_workspace
Main LINZA workflow facade. Choose one action to inspect the workspace, ingest artifacts, review/apply supervised items, explain connections, search memory, export context, or run diagnostics. Safe actions are read-only; apply actions preview by default with dry_run=true.