// SCAN REPORT

Mementos

125 tools. 59 can modify or destroy data without limits.

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

Last updated:

59 can modify or destroy data
66 read-only
125 tools total

Community server · catalogue entry verified 01/07/2026

How to control Mementos ↓

TOOL MAP

What Mementos exposes to your agents

Read (66) Write / Execute (48) Destructive / Financial (11)

Running more than one server? Check your whole stack →

MOST DANGEROUS TOOLS
Critical Risk

The most dangerous Mementos tools

Destructive · High bulk_forget This tool permanently deletes memories without apparent recovery mechanism. Destructive · High clean_expired This tool permanently deletes records from a database without the ability to recover them. Destructive · High clean_expired_locks The tool deletes resource locks, which is an irreversible operation that cannot be undone. Destructive · High entity_delete This tool permanently removes entities from the memory system with no undo capability.

59 of Mementos's 125 tools can modify, destroy, or commit something on every call — and an agent calls them with no built-in limits.

POLICY

How to control Mementos

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

Deny destructive operations
{
  "bulk_forget": {
    "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
{
  "bulk_update": {
    "limits": [
      {
        "counter": "bulk_update_per_hour",
        "window": "hour",
        "max": 30,
        "scope": "grant"
      }
    ]
  }
}

Prevents bulk unintended modifications from agents caught in loops.

Cap read operations
{
  "describe_tools": {
    "limits": [
      {
        "counter": "describe_tools_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 Mementos — 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 MEMENTOS →

Instant setup, no code required.

FULL CATALOGUE

All 125 Mementos tools

DESTRUCTIVE 11 tools
Destructive bulk_forget Delete multiple memories by IDs Destructive clean_expired Remove expired memories from the database Destructive clean_expired_locks Delete all expired resource locks. Notifies holding agents via conversations DM. Destructive entity_delete Delete an entity by name or ID. Destructive memory_consolidate Consolidate memories: deduplicate/merge near duplicates, promote repeated episodic observations into semantic Destructive memory_forget Delete a memory by ID or key Destructive memory_gdpr_erase GDPR right to be forgotten: erase all memories containing a PII identifier. Replaces content with [REDACTED], Destructive memory_synthesis_rollback Roll back a synthesis run, reversing all applied proposals. Destructive memory_unsubscribe Remove a memory subscription by ID. Destructive relation_delete Delete a relation by ID. Destructive webhook_delete Delete a persisted webhook hook by ID.
EXECUTE 3 tools
Execute build_file_dep_graph Scan a codebase directory and build a file dependency graph: creates Execute memory_auto_process Enqueue a conversation turn or text for async LLM memory extraction. Returns immediately (non-blocking). Set a Execute migrate_pg Apply PostgreSQL schema migrations to the configured RDS instance
WRITE 45 tools
Write bulk_update Update multiple memories with the same changes Write entity_create Create a knowledge graph entity (person, project, tool, concept, file, api, pattern, organization). Write entity_link Link an entity to a memory with a role (subject, object, or context). Write entity_merge Merge source entity into target. Moves all relations and memory links. Write entity_update Update an entity Write heartbeat Update agent last_seen_at to signal active session. Call periodically during long tasks to prevent being marke Write mementos_storage_feedback Save feedback for mementos Write mementos_storage_pull Pull remote PostgreSQL storage data into the local mementos database Write mementos_storage_push Push local mementos data to remote PostgreSQL storage Write mementos_storage_sync Push local changes, then pull remote changes Write memory_acl_set Set an access control rule for an agent. Patterns use for glob matching (e.g., Write memory_archive Archive a memory by ID or key (hides from lists, keeps history). No version needed. Write memory_auto_config Update auto-memory configuration at runtime (no restart needed). Set provider, model, enabled, minImportance, Write memory_autoinject_config Get or set auto-inject orchestrator config (channel-based memory push). Controls throttle, debounce, rate limi Write memory_flag Flag a memory for attention: needs-review, outdated, verify, or any custom flag. Flagged memories surface at t Write memory_import Import memories from JSON array Write memory_ingest_session Submit a session transcript for async memory extraction. Returns job_id to track progress. Write memory_invalidate Invalidate an existing fact by setting valid_until to now. Use when a contradiction is confirmed and the old f Write memory_lock Acquire an exclusive write lock on a memory key to prevent concurrent writes. Write memory_pin Pin or unpin a memory by ID or key. No version needed. Write memory_rate Rate a memory as useful or not useful. Provides feedback for memory quality tracking. Write memory_reflect Reflect on a session, task, or range with an LLM-judge critic and write structured lessons back as linked memo Write memory_save Save/upsert a memory. scope: global=all agents, shared=project, private=single agent, working=transient sessio Write memory_save_image Save an image memory. If OPENAI_API_KEY is set and image_url provided, auto-extracts a description via GPT-4o- Write memory_save_tool_event Record a tool call event (success/failure, latency, tokens). Optionally saves a lesson as a shared memory. Write memory_sync_pull Pull memories from a remote mementos-serve instance into local DB. Set MEMENTOS_REMOTE_URL or pass url. Write memory_sync_push Push local memories to a remote mementos-serve instance. Set MEMENTOS_REMOTE_URL or pass url. Write memory_synthesize Run ALMA synthesis: analyze memory corpus, find redundancies, propose and apply consolidations. Write memory_unlock Release a memory write lock. Write memory_update Update a memory Write register_agent Register an agent. Idempotent — same name returns existing agent. Write register_machine Register the current machine in the mementos machine registry. Auto-detects hostname. Idempotent by hostname. Write register_project Register a project for memory scoping Write relation_create Create a relation between two entities (uses, knows, depends_on, created_by, related_to, contradicts, part_of, Write rename_machine Rename a machine by its ID or current name. Write resource_lock Acquire a lock on any resource (project, memory, entity, agent, connector). Write resource_unlock Release a resource lock. Write send_feedback Send feedback about this service Write session_extract Extract memories from a session summary. Auto-creates structured memories from title, topics, notes. Write set_focus Set focus for an agent on a project. Memory ops will auto-scope to that project Write set_primary_machine Mark a machine as the primary machine. Only one primary machine is allowed at a time. Write unfocus Remove focus for an agent (clears project scoping). Write update_agent Update agent name, description, role, metadata, or active_project_id. Write webhook_create Create a persistent HTTP webhook hook. The URL will be POSTed with the hook context as JSON. Write webhook_update Enable, disable, or update a persisted webhook hook.
READ 66 tools
Read describe_tools Get full parameter schemas and examples for tools. Omit names to list all tools. Read entity_disambiguate Find potential duplicate entities by name similarity (trigram). Returns pairs above the threshold within same Read entity_get Get entity details by name or ID, including relations summary and memory count. Read entity_list List entities. Optional filters: type, project_id, search, limit. Read entity_unlink Unlink a memory from an entity. Read get_agent Get agent details by ID or name Read get_focus Get the current focus project for an agent. Read get_project Get a project by ID, path, or name. Read graph_path Find shortest path between two entities in the knowledge graph. Read graph_query Traverse the knowledge graph from an entity up to N hops. Returns entities and relations. Read graph_stats Comprehensive knowledge graph statistics: entity/relation counts by type, most-connected entities, orphan coun Read graph_traverse Multi-hop graph traversal from an entity. Returns all paths with entities and relations at each hop. Supports Read hook_list List all registered hooks in the in-memory registry (built-in + webhooks). Read hook_stats Get statistics about registered hooks: total, by type, blocking vs non-blocking. Read list_agent_locks List all active resource locks held by an agent. Read list_agents List all registered agents Read list_agents_by_project List agents currently active on a project. Read list_machines List all registered machines with their hostname, platform, primary status, and last seen time. Read list_projects List all registered projects Read mementos_storage_status Show mementos local database and remote storage sync status Read memory_acl_list List access control rules for an agent. Read memory_activity Get daily memory creation activity over N days. Read memory_audit Review low-trust memories (trust_score < threshold). Returns memories flagged by the poisoning detection heuri Read memory_audit_export Export the full immutable audit log for compliance reporting. Supports date range and operation type filtering Read memory_audit_trail Get the immutable audit trail for a specific memory. Shows all create/update/delete operations with timestamps Read memory_auto_status Get the current auto-memory extraction queue stats (pending, processing, processed, failed, dropped). Read memory_auto_test Test memory extraction on text WITHOUT saving anything. Returns what would be extracted. Useful for tuning pro Read memory_autoinject_status Get auto-inject orchestrator status: running state, session watcher, push history, rate limit counters, and fu Read memory_autoinject_test Test what memories would be activated by a given context WITHOUT pushing. Shows what the auto-inject pipeline Read memory_briefing Lightweight delta briefing: what memories changed since an agent Read memory_chain_get Retrieve an ordered memory chain/sequence by group ID. Returns all steps in order. Read memory_check_contradiction Check if a new memory would contradict existing high-importance facts. Call before saving to detect conflicts. Read memory_check_lock Check if a memory key is currently write-locked. Read memory_compress Compress multiple memories into a single summary memory. Uses LLM if available, otherwise truncates. Read memory_context Get memories relevant to current context. Uses time-weighted scoring: score = importance × decay(age). Pinned Read memory_context_layered Structured multi-section memory context: Core Facts, Recent History, Relevant Knowledge, Active Decisions. Bet Read memory_diff Show what changed between two versions of a memory. Compares value, importance, scope. Omit v1/v2 to diff the Read memory_evict Enforce memory bounds per scope. Archives lowest-utility memories (using decay score) when any scope exceeds i Read memory_export Export memories. format= Read memory_get Get a single memory by ID. Read memory_health Comprehensive health check for memories. Detects: stale (old + 0 access), high-importance-forgotten (importanc Read memory_inject Get memory context for system prompt injection. Selects by scope, importance, recency. Use strategy= Read memory_list List memories. Default: compact lines. full=true for complete JSON objects. Read memory_profile Synthesize a coherent profile from preference and fact memories using LLM. Cached for 24h, auto-refreshed when Read memory_recall Recall a memory by key. Returns the best matching active memory. Use as_of for temporal queries (what was true Read memory_recall_deep Deep memory recall using ASMR (Agentic Search and Memory Retrieval) — runs 3 parallel search agents (facts, co Read memory_report Get a rich summary report: totals, activity trend, top memories, scope/category breakdown. Read memory_search Search memories by keyword across key, value, summary, and tags Read memory_search_bm25 Search memories using FTS5 BM25 ranking. Returns results scored by term frequency, document length, and field Read memory_search_hybrid Hybrid search combining keyword (FTS5) and semantic (embedding) search using Reciprocal Rank Fusion (RRF). Bes Read memory_search_semantic Semantic (meaning-based) memory search using vector embeddings. Finds memories by conceptual similarity, not k Read memory_session_list List session memory extraction jobs. Read memory_session_status Get the status of a session memory extraction job. Read memory_stale Find memories not accessed recently. Useful for cleanup or review. Read memory_stats Get aggregate statistics about stored memories Read memory_subscribe Subscribe an agent to memory change notifications. Matches by key pattern (glob) and/or tag pattern. Read memory_sync_status Check if a remote mementos-serve is reachable. Set MEMENTOS_REMOTE_URL or pass url. Read memory_synthesis_history List past synthesis runs. Read memory_synthesis_status Get the status of synthesis runs. Read memory_tool_insights Get usage stats, lessons learned, and recommendations for MCP tools. Helps agents avoid past mistakes and reus Read memory_versions Get version history for a memory. Shows what changed across updates. Read relation_get Get a relation by ID. Read relation_list List relations for an entity. Filter by type and direction (outgoing, incoming, both). Read resource_check_lock Check active locks on a resource. Read search_tools Search available tools by name or keyword. Returns names only. Read webhook_list List all persisted webhook hooks.
EXPLORE

Related servers

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

Yaver 1060 tools
adminautomation
Nodebench 824 tools
adminautomation
UnClick 813 tools
adminautomation
Binance MCP Server 734 tools
adminautomation
Fortimanager 584 tools
adminautomation
Crow 573 tools
adminautomation
Mcp 571 tools
adminautomation
GoHighLevel MCP Server 566 tools
adminautomation
FAQ

Questions about Mementos

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

Yes. The Mementos server exposes 11 destructive tools including bulk_forget, clean_expired, clean_expired_locks. 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 Mementos? +

The Mementos server has 45 write tools including bulk_update, entity_create, entity_link. 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 Mementos.

How many tools does the Mementos MCP server expose? +

125 tools across 4 categories: Destructive, Execute, Read, Write. 66 are read-only. 59 can modify, create, or delete data.

How do I enforce a policy on Mementos? +

Register the Mementos 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 Mementos tool call.

Deterministic rules across all 125 Mementos tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.

ENFORCE POLICY ON MEMENTOS →

Instant setup, no code required.

125 Mementos 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.