Memory MCP Server — Persistent Agent Memory

Memory MCP Server providing persistent memory storage for AI agents — storing facts, preferences, conversation summaries, and contextual information across sessions, enabling agents to remember user preferences, past conversations, and learned information beyond the context window through structured memory operations.

Evaluated Mar 06, 2026 (0d ago) vcurrent
Homepage ↗ Repo ↗ Developer Tools memory persistence mcp-server agent-memory knowledge-base context
⚙ Agent Friendliness
76
/ 100
Can an agent use this?
🔒 Security
76
/ 100
Is it safe for agents?
⚡ Reliability
67
/ 100
Does it work consistently?

Score Breakdown

⚙ Agent Friendliness

MCP Quality
68
Documentation
70
Error Messages
68
Auth Simplicity
95
Rate Limits
90

🔒 Security

TLS Enforcement
80
Auth Strength
72
Scope Granularity
72
Dep. Hygiene
70
Secret Handling
85

Local storage. No credentials. Community MCP. Protect stored memory. Don't store secrets in memory.

⚡ Reliability

Uptime/SLA
72
Version Stability
65
Breaking Changes
62
Error Recovery
68
AF Security Reliability

Best When

An AI agent needs to persist facts, preferences, and summaries across multiple sessions — providing memory continuity beyond the LLM context window.

Avoid When

You need full-text search, vector similarity search, or multi-user memory — use ChromaDB, Neo4j memory, or LanceDB MCPs.

Use Cases

  • Storing and retrieving user preferences from personalization agents
  • Persisting conversation history summaries across sessions from assistant agents
  • Building user-specific knowledge bases from personal assistant agents
  • Remembering project context across separate coding sessions from development agents
  • Accumulating learned facts about users and tasks from continuous learning agents
  • Creating agent memory that survives context window resets

Not For

  • High-volume data storage (this is agent context memory, not a database)
  • Shared memory across multiple users (single-user memory store)
  • Teams with specific vector search requirements (use vector DB MCPs)

Interface

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

Authentication

Methods: none
OAuth: No Scopes: No

No external authentication — local SQLite or file-based storage. Data stored locally on the machine.

Pricing

Model: free
Free tier: Yes
Requires CC: No

Free, open source community MCP.

Agent Metadata

Pagination
none
Idempotent
Full
Retry Guidance
Not documented

Known Gotchas

  • Memory grows unboundedly without pruning — implement memory management strategies
  • No semantic search — retrieval is keyword or key-based, not vector similarity
  • Data is local machine only — not shared across machines or users
  • Memory structure and schema should be designed before heavy use
  • Community MCP — storage format may change between versions
  • Agent may confuse old memories with current context — recency weighting important

Alternatives

Full Evaluation Report

Comprehensive deep-dive: security analysis, reliability audit, agent experience review, cost modeling, competitive positioning, and improvement roadmap for Memory MCP Server — Persistent Agent Memory.

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 2026-03-06.

5647
Packages Evaluated
26151
Need Evaluation
173
Need Re-evaluation
Community Powered