neo4j-agent-memory

⚠️ Verify authoritative state before writing. Version numbers, extras, tool counts, and API surface change between releases. The values in this skill reflect a specific point in time. Before publishing anything version-sensitive, confirm against PyPI (

https://pypi.org/project/neo4j-agent-memory/

) and the GitHub README (

https://github.com/neo4j-labs/agent-memory

). PyPI is the authoritative source for version numbers — never infer.

• Building AI agents that need persistent memory (short-term, long-term, reasoning traces) backed by Neo4j
• Using the

  neo4j-agent-memory

  Python package or the hosted NAMS service at memory.neo4jlabs.com
• Integrating agent memory with LangChain, PydanticAI, CrewAI, AWS Strands, Google ADK, OpenAI Agents, LlamaIndex, or Microsoft Agent Framework
• Writing documentation, tutorials, or positioning content about graph-native agent memory
• Comparing graph-native memory against vector-only approaches

• Plain Neo4j driver connections (no memory layer needed) → use

  neo4j-driver-python-skill

• Writing or optimizing Cypher queries → use

  neo4j-cypher-skill

• GraphRAG retrieval pipelines → use

  neo4j-graphrag-skill

neo4j-agent-memory

pip install neo4j-agent-memory

https://memory.neo4jlabs.com

⚠️ NAMS is reachable but not yet referenced in the GitHub README or

neo4j.com/labs/agent-memory/

. Treat it as early-access / soft-launched. Do not assert SLAs, pricing, or GA status in published content. See the Hosted Service (NAMS) section below for details.

• Person
• Organization
• Location
• Event
• +O Object (anything that doesn't fit the core four — products, concepts, projects, etc.)

{
  "mcpServers": {
    "neo4j-agent-memory": {
      "command": "uvx",
      "args": ["neo4j-agent-memory[mcp]", "mcp", "serve", "--password", "your-password"],
      "env": {
        "OPENAI_API_KEY": "sk-..."
      }
    }
  }
}

For the hosted MCP endpoint at

memory.neo4jlabs.com/mcp

, see the Hosted Service (NAMS) section below — it uses SSE transport and bearer-token auth, not a local

uvx

invocation.

neo4j-agent-memory

https://memory.neo4jlabs.com

⚠️ Verify against the live service before citing. NAMS is not documented on the GitHub README or

neo4j.com/labs/agent-memory/

. Endpoint shapes, tool counts, auth flows, and limits can change without a release note. Before publishing anything NAMS-specific, re-check the live site and the OpenAPI spec at

/openapi.json

.

• Base URL:

  https://memory.neo4jlabs.com

• Web console: root URL — workspace management, memory browsing, entity visualization
• REST API:

  https://memory.neo4jlabs.com/v1/

  — OpenAPI spec at

  /openapi.json

  ; covers conversations, entities, observations, reasoning traces, and read-only Cypher
• MCP endpoint:

  https://memory.neo4jlabs.com/mcp

  — SSE transport, exposes the hosted tool set, bearer-token auth

• API keys, prefixed

  nams_

  , created and rotated from the web console — used as a bearer token for REST and MCP
• Auth0 OAuth2 (PKCE) + scoped JWTs for interactive user flows

--password

1. spaCy — fast statistical NER, cheapest, broad but imprecise coverage
2. GLiNER — zero-shot entity extraction with typed schemas;

   GLiREL

   for relationships
3. LLM fallback — most accurate, most expensive; used when structure is rich or ambiguous

DeduplicationConfig

ExtractionConfig

DeduplicationConfig

MemoryIntegration

SessionStrategy

• "Graph handles understanding; vector handles similarity."
• "Vector stores give you recall. The graph gives you understanding."
• "Three memory types, one knowledge graph."

• Anchor to the Foundation Capital "AI's Trillion-Dollar Opportunity: Context Graphs" thesis when the forum warrants it.

• neo4j-agent-memory

  is positioned as the complete implementation of the context graph category — it covers all three memory layers, not just two.
• The context graph coexists with domain data in the same Neo4j instance (not a bolted-on external system). This is a key conceptual/visual point for architecture diagrams.

• "graph-native memory"
• "context graph"
• "three distinct memory layers"
• "reasoning traces as first-class graph nodes"
• "learn from past reasoning"
• "build knowledge graphs automatically from conversations"
• "Neo4j Labs project" / "experimental" / "community-supported"

• Don't name specific competitors (Mem0, Zep, Letta, Cognee, Supermemory) in published content. Reframe comparisons around capabilities, not product names.
• Don't call it "production-ready" (it's a Labs project — see the

  neo4j-labs-brand

  skill for the full voice guide).
• Don't say "officially supported" or imply SLAs.

1. Outdated version numbers — anyone writing "v0.1.0" today may be working from stale notes; verify PyPI.
2. Wrong canonical docs URL — it's

   neo4j.com/labs/agent-memory

   , not a Vercel preview URL.
3. Inferred API surface — if code samples weren't run, flag them; prefer patterns from the GitHub README or official examples.
4. Missing "Labs" framing — experimental/community-supported should be clear.
5. Conflating with other Neo4j MCP servers — there are several (

   mcp-neo4j-cypher

   ,

   mcp-neo4j-memory

   — the old knowledge graph memory server, etc.).

   neo4j-agent-memory

   's MCP server is distinct and ships as part of the package under the

   [mcp]

   extra.
6. Confusing NAMS with the self-hosted library — same underlying project, different consumption models. Connection strings, auth, and tool sets differ: self-hosted uses a local

   uvx

   invocation and a Neo4j

   --password

   ; NAMS uses an SSE MCP URL and a

   nams_

   -prefixed bearer token. Don't mix them.
7. Over-promising NAMS availability — the hosted service is not yet referenced in the GitHub README or

   neo4j.com/labs/agent-memory/

   . Avoid "officially supported," SLAs, pricing claims, or "production-ready" framing. Treat it as early-access.

• examples/lennys-memory

  — flagship PydanticAI demo

• examples/full-stack-chat-agent

  — PydanticAI news research with NVL graph viz

• examples/financial-services-advisor/aws-financial-services-advisor

  — AWS Strands multi-agent KYC/AML with reasoning-trace audit trails

• examples/financial-services-advisor/google-cloud-financial-advisor

  — Google ADK multi-agent with Vertex AI embeddings + SSE streaming

• examples/microsoft_agent_retail_assistant

  — Microsoft Agent Framework with GDS algorithms and context providers

• examples/domain-schemas

  — 8 GLiNER2 extraction scripts

• examples/google_cloud_integration

  — progressive tutorial covering Vertex AI → ADK → MCP →

  MemoryIntegration

• examples/google_adk_demo

  — standalone ADK demo of

  Neo4jMemoryService

• excalidraw

  skill — JSON format and the project's diagram management script

• neo4j-styleguide

  skill — Cypher code style and Neo4j brand colors

• neo4j-labs-brand

  skill — Labs purple (

  #6366F1

  ), status badges, disclaimer language

Short-Term:  #B2F2BB fill / #2F9E44 stroke  (green)
Long-Term:   #FFEC99 fill / #F08C00 stroke  (orange/yellow)
Reasoning:   #D0BFFF fill / #9C36B5 stroke  (purple)
Neo4j/Store: #A5D8FF fill / #1971C2 stroke  (blue)
Labs accent: #6366F1 (purple, for Labs branding elements)

neo4j.com/labs/agent-memory

diataxis

• Tutorials — build your first memory-enabled agent
• How-To Guides — entity extraction, deduplication, enrichment, integrations
• Reference — configuration, CLI, MCP tools, API
• Explanation — POLE+O model, memory types, extraction pipeline

• Version number is current per PyPI (not inferred from notes)
• Canonical docs link points to

  neo4j.com/labs/agent-memory

• Three memory types named correctly (short-term, long-term, reasoning)
• POLE+O model named consistently (not POLEO, not POLE-O)
• Python ≥ 3.10 and Neo4j ≥ 5.20 requirements are stated if relevant
• Labs disclaimer present for README/landing content
• No competitor names in published positioning
• Reasoning memory is called out as the differentiator
• Import paths use

  neo4j_agent_memory.integrations.<framework>

  (underscore, snake_case)
• If NAMS is referenced, distinguish clearly from the self-hosted library and re-verify endpoints against the live service (not yet mirrored in the README)

• PyPI (authoritative for version): https://pypi.org/project/neo4j-agent-memory/
• GitHub: https://github.com/neo4j-labs/agent-memory
• Canonical docs: https://neo4j.com/labs/agent-memory/
• CHANGELOG: https://github.com/neo4j-labs/agent-memory/blob/main/CHANGELOG.md
• NAMS (hosted service): https://memory.neo4jlabs.com
• NAMS REST API: https://memory.neo4jlabs.com/v1/ (OpenAPI at

  /openapi.json

  )
• NAMS MCP endpoint: https://memory.neo4jlabs.com/mcp (SSE)
• create-context-graph (scaffolder): https://create-context-graph.dev
• Community Forum: https://community.neo4j.com
• Microsoft Learn integration page: https://learn.microsoft.com/en-us/agent-framework/integrations/neo4j-memory

• Version: check PyPI before citing
• Consumption model: self-hosted vs NAMS
• Correct extras installed (

  neo4j-agent-memory[<extra>]

  )

• MemoryClient

  as async context manager
• Three types named: short-term / long-term / reasoning
• POLE+O consistent (not POLEO or POLE-O)
• NAMS: early-access framing; no SLAs/pricing
• Credentials not hardcoded; NAMS bearer token separate from Neo4j

  --password