// SCAN REPORT

Cairntir

7 tools. 3 can modify or destroy data without limits.

3 write tools that can modify data. Rate limits recommended.

Last updated: 3 Jul 2026

3 can modify or destroy data

4 read-only

7 tools total

Community server · catalogue entry verified 03/07/2026

How to control Cairntir ↓

TOOL MAP

What Cairntir exposes to your agents

Read (4) Write / Execute (3) Destructive / Financial (0)

Running more than one server? Check your whole stack →

MOST DANGEROUS TOOLS

High Risk

The most dangerous Cairntir tools

Execute · Medium cairntir_audit This tool triggers execution of a Quality skill against a wing (likely a data partition in the SQLite memory system). Execute · Medium cairntir_crucible The tool appears to execute a reasoning or evaluation process ('stress-test') against a provided claim. Write · Medium cairntir_remember This tool writes new memory entries into the local SQLite database.

3 of Cairntir's 7 tools can modify, destroy, or commit something on every call — and an agent calls them with no built-in limits.

POLICY

How to control Cairntir

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

Rate limit write operations

{
  "cairntir_remember": {
    "limits": [
      {
        "counter": "cairntir_remember_per_hour",
        "window": "hour",
        "max": 30,
        "scope": "grant"
      }
    ]
  }
}

Prevents bulk unintended modifications from agents caught in loops.

Cap read operations

{
  "cairntir_cross_recall": {
    "limits": [
      {
        "counter": "cairntir_cross_recall_per_minute",
        "window": "minute",
        "max": 60,
        "scope": "grant"
      }
    ]
  }
}

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

Create a free account and register Cairntir — nothing to install.
Add these rules — paste them, or build them visually. Tune the limits to your setup.
Point your MCP client (Claude, Cursor, anything) at your gateway URL.

ENFORCE POLICY ON CAIRNTIR →

Instant setup, no code required.

FULL CATALOGUE

All 7 Cairntir tools

EXECUTE 2 tools

Execute cairntir_audit Run the Quality skill over a wing. Execute cairntir_crucible Stress-test a claim with the Crucible skill.

WRITE 1 tools

Write cairntir_remember Store a verbatim memory drawer in a wing/room.

READ 4 tools

Read cairntir_cross_recall Semantic search across EVERY wing. Use when a question might find Read cairntir_recall Semantic search across stored drawers. Read cairntir_session_start Load the 4-layer context for a wing (the amnesia killer). Read cairntir_timeline Chronological view of drawers in a wing mentioning an entity.

EXPLORE

Related servers

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

FAQ

Questions about Cairntir

How do I prevent bulk modifications through Cairntir? +

The Cairntir server has 1 write tools including cairntir_remember. 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 Cairntir.

How many tools does the Cairntir MCP server expose? +

7 tools across 2 categories: Execute, Read. 4 are read-only. 3 can modify, create, or delete data.

How do I enforce a policy on Cairntir? +

Register the Cairntir 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 Cairntir tool call.

Deterministic rules across all 7 Cairntir tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.

ENFORCE POLICY ON CAIRNTIR →

Instant setup, no code required.

7 Cairntir tools catalogued and risk-classified — across an index of 43,000+ MCP servers.