Obsidian Palace MCP

31 tools. 13 can modify or destroy data without limits.

2 destructive tools with no built-in limits. Policy required.

Last updated:

13 can modify or destroy data
18 read-only
31 tools total

Community server · catalogue entry verified 04/07/2026

How to control Obsidian Palace MCP ↓

What Obsidian Palace MCP exposes to your agents

Read (18) Write / Execute (11) Destructive / Financial (2)
Critical Risk

The most dangerous Obsidian Palace MCP tools

13 of Obsidian Palace MCP's 31 tools can modify, destroy, or commit something on every call — and an agent calls them with no built-in limits.

How to control Obsidian Palace MCP

PolicyLayer is an MCP gateway — it sits between your AI agents and Obsidian Palace MCP, and nothing reaches the server without passing your rules. These are the rules we recommend:

Deny destructive operations
{
  "palace_batch": {
    "deny_if": [
      {
        "conditions": [],
        "on_deny": "Blocked by default. Requires approval."
      }
    ]
  }
}

Destructive tools should never be available to autonomous agents without human approval.

Rate limit write operations
{
  "palace_autolink": {
    "limits": [
      {
        "counter": "palace_autolink_per_hour",
        "window": "hour",
        "max": 30,
        "scope": "grant"
      }
    ]
  }
}

Prevents bulk unintended modifications from agents caught in loops.

Cap read operations
{
  "palace_check": {
    "limits": [
      {
        "counter": "palace_check_per_minute",
        "window": "minute",
        "max": 60,
        "scope": "grant"
      }
    ]
  }
}

Controls API costs and prevents retry loops from exhausting upstream rate limits.

  1. Create a free account and register Obsidian Palace MCP — nothing to install.
  2. Add these rules — paste them, or build them visually. Tune the limits to your setup.
  3. Point your MCP client (Claude, Cursor, anything) at your gateway URL.
ENFORCE POLICY ON OBSIDIAN PALACE →

Instant setup, no code required.

All 31 Obsidian Palace MCP tools

READ 18 tools
Read palace_check Check for existing knowledge before creating new notes. Returns matches, domain suggestions, and a recommendat Read palace_clarify Detect context and generate clarifying questions before storing knowledge. Use this when palace_store intent Read palace_export Export notes in various formats. Supports single notes, hub + children consolidated, or directories. Formats: Read palace_history View version history for a note. Shows timestamps, operations, and optionally diffs between versions. Use comp Read palace_links Get incoming links (backlinks) and/or outgoing links for a note. Supports multi-hop traversal with depth param Read palace_list List notes in a palace directory. Can list recursively and include metadata. Read palace_orphans Find orphan notes and optionally clean them up. Orphan Types: - Read palace_project_summary Load project context for AI session resume. Returns project status, work items, recent changes, time tracking, Read palace_query Query notes by properties like type, tags, confidence, verified status, and dates. Use this for filtering with Read palace_read Read a specific note from the palace by path or title. Returns the full note content and metadata. Read palace_recall Search the Obsidian palace for knowledge. Uses FTS5 full-text search for fast, ranked results. Supports cross- Read palace_related Find notes related to a given note based on shared links, shared tags, or both. Returns a ranked list of relat Read palace_standards Load binding standards that AI should follow. Returns standards notes with ai_binding frontmatter. Standards Read palace_standards_validate Validate a note against applicable standards. Returns a compliance report with violations and warnings. Read palace_structure Get the palace directory tree structure with domain pattern analysis. Essential for understanding vault organi Read palace_stubs List all stub notes (placeholders) in a vault that need expansion. Use this tool to: 1. Find stubs that need Read palace_time_summary Aggregate and report on logged time entries. Filter by project, client, category, date range, or billable stat Read palace_vaults List all configured vaults with their aliases, paths, access modes, and optionally note counts.

Related servers

Other MCP servers with similar tools — same risk classification, starter policies for each.

Questions about Obsidian Palace MCP

Can an AI agent delete data through the Obsidian Palace MCP server? +

Yes. The Obsidian Palace MCP server exposes 2 destructive tools including palace_batch, palace_delete. These permanently remove resources with no undo. PolicyLayer blocks destructive tools by default so they never reach the upstream server.

How do I prevent bulk modifications through Obsidian Palace MCP? +

The Obsidian Palace MCP server has 10 write tools including palace_autolink, palace_improve, palace_migrate. Set a rate limit in your policy -- for example, 10 calls per hour prevents an agent from making more than 10 modifications per hour. PolicyLayer enforces this at the gateway, before calls reach Obsidian Palace MCP.

How many tools does the Obsidian Palace MCP server expose? +

31 tools across 4 categories: Destructive, Execute, Read, Write. 18 are read-only. 13 can modify, create, or delete data.

How do I enforce a policy on Obsidian Palace MCP? +

Register the Obsidian Palace MCP server in PolicyLayer, apply the suggested rules above (adjust the limits to your use case), and point your AI client at the PolicyLayer proxy URL instead of the server directly. Your agents keep the same tools; PolicyLayer evaluates every call against policy before it executes. Nothing to install, live in minutes.

Enforce policy on every Obsidian Palace MCP tool call.

Deterministic rules across all 31 Obsidian Palace MCP tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.

Instant setup, no code required.

31 Obsidian Palace MCP tools catalogued and risk-classified — across an index of 43,000+ MCP servers.

// GET IN TOUCH

Have a question or want to learn more? Send us a message.

Message sent.

We'll get back to you soon.