// SCAN REPORT

Graph-Memory

23 tools. 11 can modify or destroy data without limits.

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

Last updated: 30 Jun 2026

11 can modify or destroy data

12 read-only

23 tools total

Community server · catalogue entry verified 30/06/2026

How to control Graph-Memory ↓

TOOL MAP

What Graph-Memory exposes to your agents

Read (12) Write / Execute (8) Destructive / Financial (3)

Running more than one server? Check your whole stack →

MOST DANGEROUS TOOLS

Critical Risk

The most dangerous Graph-Memory tools

Destructive · High graph_decay This tool irreversibly modifies node confidence values and edge weights across the knowledge graph without the ability to undo the changes directly (requiring external backup restoration). Destructive · High graph_delete This tool irreversibly removes data from the knowledge graph. Destructive · High graph_prune This tool permanently deletes entities and edges from the user's personal knowledge graph based on decay thresholds. Execute · High graph_cypher Although the tool enforces read-only constraints via Neo4j executeRead(), it allows the AI to generate and execute arbitrary Cypher queries against a personal knowledge graph.

11 of Graph-Memory's 23 tools can modify, destroy, or commit something on every call — and an agent calls them with no built-in limits.

POLICY

How to control Graph-Memory

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

Deny destructive operations

{
  "graph_decay": {
    "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

{
  "graph_audit": {
    "limits": [
      {
        "counter": "graph_audit_per_hour",
        "window": "hour",
        "max": 30,
        "scope": "grant"
      }
    ]
  }
}

Prevents bulk unintended modifications from agents caught in loops.

Cap read operations

{
  "graph_build_context": {
    "limits": [
      {
        "counter": "graph_build_context_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 Graph-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 GRAPH-MEMORY →

Instant setup, no code required.

FULL CATALOGUE

All 23 Graph-Memory tools

DESTRUCTIVE 3 tools

Destructive graph_decay Apply time-based decay to every node confidence and edge weight using per-type half-lives (preferences ~693d, Destructive graph_delete Permanently delete an entity node and all its edges by ID. Use for removing duplicate or erroneous nodes. Cann Destructive graph_prune Remove entities and edges that have decayed below threshold. DESTRUCTIVE — always preview first. Requires user

EXECUTE 1 tools

Execute graph_cypher Execute a read-only Cypher query against the memory graph. You generate the Cypher — this tool just runs it. E

WRITE 7 tools

Write graph_audit Append a structured event to the dream process audit log (logs/dream-audit.jsonl). Write graph_boost Increase an edge Write graph_ingest Queue a document for asynchronous extraction into the memory graph (mode= Write graph_merge Consolidate two entities into one — moves source Write graph_relate Create or strengthen a relationship between entities. Creates the endpoint entities if they don Write graph_unmerge Split a falsely merged entity back into two separate entities, redistributing specified edges. Use when entity Write graph_weaken Decrease an edge

READ 12 tools

Read graph_build_context Single tool call that bundles a session Read graph_communities Find clusters of densely-interconnected entities in the graph. Uses greedy seed-based BFS through Read graph_contradictions Find facts that contradict each other in the memory graph — pairs connected by a CONTRADICTS edge. Use during Read graph_entities Browse or search the entity catalog. Use to check if an entity exists before creating one with graph_relate, o Read graph_export Export all graph nodes and edges to a timestamped JSONL backup file in the backups/ directory. Read graph_merge_suggestions Surface candidate pairs of entities likely to be duplicates. Read-only — never auto-merges. Combines embedding Read graph_query Query the memory graph by canonical entity name. Use when you know the entity name or close-to-canonical form Read graph_read_transcript Read and parse a Claude Code JSONL transcript file through the canonical transcript parser. Read graph_reembed Regenerate semantic-search embeddings for entities. By default only fills missing embeddings Read graph_search Find entities semantically similar to a natural-language query, then optionally expand via Read graph_stats Graph health dashboard — node/edge counts by type, average weight, orphan count, unresolved contradictions, st Read graph_validate Scan recently extracted entities and edges for quality issues: generic names, reference language,

EXPLORE

Related servers

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

FAQ

Questions about Graph-Memory

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

Yes. The Graph-Memory server exposes 3 destructive tools including graph_decay, graph_delete, graph_prune. 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 Graph-Memory? +

The Graph-Memory server has 7 write tools including graph_audit, graph_boost, graph_ingest. 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 Graph-Memory.

How many tools does the Graph-Memory MCP server expose? +

23 tools across 4 categories: Destructive, Execute, Read, Write. 12 are read-only. 11 can modify, create, or delete data.

How do I enforce a policy on Graph-Memory? +

Register the Graph-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 Graph-Memory tool call.

Deterministic rules across all 23 Graph-Memory tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.

ENFORCE POLICY ON GRAPH-MEMORY →

Instant setup, no code required.

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