memgraph-mcp-server

Provides an MCP server that lets an agent execute Cypher queries against a Memgraph database and fetch schema information, with a default block on destructive/unsafe queries unless explicitly enabled.

Evaluated Apr 04, 2026 (17d ago)
Repo ↗ Infrastructure mcp memgraph graph-database cypher agent-tools go
⚙ Agent Friendliness
60
/ 100
Can an agent use this?
🔒 Security
41
/ 100
Is it safe for agents?
⚡ Reliability
24
/ 100
Does it work consistently?

Score Breakdown

⚙ Agent Friendliness

MCP Quality
75
Documentation
70
Error Messages
0
Auth Simplicity
85
Rate Limits
10

🔒 Security

TLS Enforcement
30
Auth Strength
55
Scope Granularity
20
Dep. Hygiene
35
Secret Handling
60

Mentions blocking destructive queries (DELETE/REMOVE/DROP/CREATE/MERGE/SET) by default unless unsafe: true is provided to run_query—this is a positive safety control. However, TLS enforcement and transport security are not documented in the provided content; auth is limited to Memgraph user/password (or none by default). There is no mention of fine-grained authorization scopes for tools or request-level auditing. Rate limits and robust operational protections are not described.

⚡ Reliability

Uptime/SLA
0
Version Stability
40
Breaking Changes
20
Error Recovery
35
AF Security Reliability

Best When

You want an MCP-compatible interface for graph querying and can control/limit agent permissions and query safety (unsafe=false by default).

Avoid When

You cannot safely control the agent’s queries or you require documented operational protections like rate limiting, audit logs, and robust error taxonomy.

Use Cases

  • Agent-driven read-only exploration of a Memgraph graph (e.g., inspect entities/relations)
  • Generating Cypher to answer questions about a graph
  • Schema discovery for prompt/tool planning via get_schema

Not For

  • Executing destructive writes/DDL in a multi-tenant or high-risk setting
  • Environments requiring strong centralized authZ/auditing beyond Memgraph credentials
  • Use without careful query constraints/approval when unsafe operations are possible

Interface

REST API
No
GraphQL
No
gRPC
No
MCP Server
Yes
SDK
No
Webhooks
No

Authentication

Methods: Memgraph username/password via environment variables or CLI flags No-auth mode via default bolt://localhost:7687 without credentials
OAuth: No Scopes: No

Authentication is delegated to Memgraph connectivity. There is no mention of per-tool auth scopes; safety is handled via blocking of unsafe Cypher unless 'unsafe: true' is used in run_query.

Pricing

Free tier: No
Requires CC: No

Self-hosted open-source tool implied; no usage pricing described in provided content.

Agent Metadata

Pagination
none
Idempotent
False
Retry Guidance
Not documented

Known Gotchas

  • run_query is capable of executing arbitrary Cypher; destructive queries require explicitly setting unsafe: true
  • Default connection may be unauthenticated (bolt://localhost:7687) if you do not provide MEMGRAPH_USER/MEMGRAPH_PASSWORD, which can be unsafe in non-local environments
  • No documentation is provided here for rate limiting, timeouts, or retry/idempotency guidance—agents may need to implement their own safeguards

Alternatives

Full Evaluation Report

Comprehensive deep-dive: security analysis, reliability audit, agent experience review, cost modeling, competitive positioning, and improvement roadmap for memgraph-mcp-server.

AI-powered analysis · PDF + markdown · Delivered within 30 minutes

$99

Package Brief

Quick verdict, integration guide, cost projections, gotchas with workarounds, and alternatives comparison.

Delivered within 10 minutes

$3

Score Monitoring

Get alerted when this package's AF, security, or reliability scores change significantly. Stay ahead of regressions.

Continuous monitoring

$3/mo

Scores are editorial opinions as of 2026-04-04.

8642
Packages Evaluated
17761
Need Evaluation
586
Need Re-evaluation
Community Powered