// SCAN REPORT

Munin Memory

21 tools. 6 can modify or destroy data without limits.

1 destructive tool with no built-in limits. Policy required.

Last updated: 29 Jun 2026

6 can modify or destroy data

15 read-only

21 tools total

Community server · catalogue entry verified 29/06/2026

How to control Munin Memory ↓

TOOL MAP

What Munin Memory exposes to your agents

Read (15) Write / Execute (5) Destructive / Financial (1)

Running more than one server? Check your whole stack →

MOST DANGEROUS TOOLS

Critical Risk

The most dangerous Munin Memory tools

Destructive · High memory_delete The tool permanently removes data from user-controlled storage with no undo mechanism. Write · Medium memory_consolidate This tool processes and merges/synthesizes existing log entries into consolidated memory records. Write · Medium memory_log The tool appends (creates) new log entries to a persistent database. Write · Medium memory_update_status This tool creates or modifies data (status entries) in a controlled, self-hosted database.

6 of Munin Memory's 21 tools can modify, destroy, or commit something on every call — and an agent calls them with no built-in limits.

POLICY

How to control Munin Memory

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

Deny destructive operations

{
  "memory_delete": {
    "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

{
  "memory_consolidate": {
    "limits": [
      {
        "counter": "memory_consolidate_per_hour",
        "window": "hour",
        "max": 30,
        "scope": "grant"
      }
    ]
  }
}

Prevents bulk unintended modifications from agents caught in loops.

Cap read operations

{
  "memory_attention": {
    "limits": [
      {
        "counter": "memory_attention_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 Munin Memory — 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 MUNIN MEMORY →

Instant setup, no code required.

FULL CATALOGUE

All 21 Munin Memory tools

DESTRUCTIVE 1 tools

Destructive memory_delete Delete a specific state entry by namespace+key, or all entries in a namespace. First call without delete_token

WRITE 5 tools

Write memory_consolidate Manually trigger memory consolidation for a specific namespace or all eligible tracked namespaces. Consolidati Write memory_log Append a chronological log entry. Log entries are immutable and timestamped. Use for decisions, events, and mi Write memory_retrieval_feedback Submit explicit feedback on retrieval quality. Use after a search/query returned poor results, missed an expec Write memory_update_status Update a tracked status entry in Write memory_write Store or update a state entry in memory. If an entry with the same namespace+key exists, it will be overwritte

READ 15 tools

Read memory_attention Return deterministic triage items for tracked work. Surfaces blocked statuses, stale active work, expiring or Read memory_commitments Surface explicit commitments derived from tracked next steps and dated, attributable source text. Use this whe Read memory_extract Suggest reviewable memory operations from explicit conversation signals. Use this after Read memory_get Retrieve the full content of a single memory entry by its UUID. Use this after Read memory_handoff Assemble a source-backed handoff pack for one namespace: current state, recent decisions, open loops, recent a Read memory_history View the chronological audit trail of changes to memory. Returns a timeline of writes, updates, deletes, names Read memory_insights Return per-entry retrieval analytics: how often each entry was retrieved (impressions), opened (opens), follow Read memory_list Browse memory contents. Without a namespace: shows all namespaces with entry counts and last_activity_at (demo Read memory_narrative Derive a compact narrative view for one namespace from current status, recent logs, and audit history. Use thi Read memory_patterns Derive conservative, reviewable patterns from repeated decision logs, tracked-status follow-through, and commi Read memory_query Search and filter memories. Supports lexical (keyword), semantic (vector similarity), and hybrid (RRF fusion o Read memory_read Retrieve a specific state entry by namespace and key. Use this when you already know both. Returns the full co Read memory_read_batch Retrieve multiple state entries in a single call. Returns an array of results (found or not found) in the same Read memory_resume Build a compact, targeted continuation pack after Read memory_status Returns server capabilities, version, and feature availability. Use to discover what search modes, tools, and

EXPLORE

Related servers

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

FAQ

Questions about Munin Memory

Can an AI agent delete data through the Munin Memory MCP server? +

Yes. The Munin Memory server exposes 1 destructive tools including memory_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 Munin Memory? +

The Munin Memory server has 5 write tools including memory_consolidate, memory_log, memory_retrieval_feedback. 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 Munin Memory.

How many tools does the Munin Memory MCP server expose? +

21 tools across 4 categories: Destructive, Execute, Read, Write. 15 are read-only. 6 can modify, create, or delete data.

How do I enforce a policy on Munin Memory? +

Register the Munin Memory 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 Munin Memory tool call.

Deterministic rules across all 21 Munin Memory tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.

ENFORCE POLICY ON MUNIN MEMORY →

Instant setup, no code required.

21 Munin Memory tools catalogued and risk-classified — across an index of 43,000+ MCP servers.