Agent memory systems with A2A protocol support
Agent memory systems that support the Agent2Agent (A2A) protocol for cross-agent communication. Ranked by maturity tier.
14 systems match this filter
Tier 1 — battle-tested (3)
- Galileo (galileo.ai) (now Splunk Agent Observability) Real-time intent / belief drift detection
Acquired by Cisco (announced 2026-04-09, closed by 2026-05); now Splunk Agent Observability. Treats memory as first-class in multi-agent tracing. Luna-2 SLMs (3B / 8B) scan every interaction for intent drift and belief drift; 20+ checks …
- Google Gemini Memory Personal Context + Personal Intelligence
Gemini app learns from past chats via the Personal context setting (on by default, can be disabled; temporary chats excluded). Users can import memories and full chat history from ChatGPT, Claude and other assistants (18+, personal accou…
- Microsoft Copilot Memory Org / tenant-scoped (Exchange-backed)
Microsoft 365 Copilot memory (saved memories, chat-history details, custom instructions) stored in a hidden folder of the user's Exchange mailbox under mailbox compliance; on by default, admin-controllable; available to Copilot Chat user…
Tier 2 — production-ready (10)
- LangGraph Persistence Checkpoints + Store System
Distinct from LangMem. Built-in checkpointer saves graph state per super-step (short-term, thread-scoped). Store System adds long-term hierarchical key-value memory across threads with optional vector search + TTL. Postgres / Mongo / Red…
- Semantic Kernel Memory Memory plugin for SK agents
Microsoft Semantic Kernel's memory plugin (Microsoft.SemanticKernel.Plugins.Memory): a plug-in approach to long-term memory in the SK agent framework (agent memory marked experimental). Semantic Kernel's README names Microsoft Agent Fram…
- Vertex AI Memory Bank (now Agent Platform Memory Bank) Managed Vertex memory service
Google Cloud managed memory service, now 'Agent Platform Memory Bank' in Gemini Enterprise Agent Platform. LLM-based memory extraction/consolidation from sessions (background, multimodal, custom topics), per-identity isolation, similarit…
- Google ADK Memory SessionService tiers + VertexAIMemoryBank
Google's Agent Development Kit. Session services: InMemory / Database / VertexAI. Long-term: InMemoryMemoryService or VertexAIMemoryBankService (uses Vertex AI Memory Bank Preview for persistent cross-session memory).
- LangMem (LangChain) Episodic / semantic / procedural
First-class memory module inside LangChain — explicit cognitive-science-style typing of memories.
- LlamaIndex Memory FIFO short-term + Memory Blocks
Two-tier model: ChatMessage FIFO queue with overflow flushed into long-term Memory Block objects. Memory class replacing older ChatMemoryBuffer.
- Mastra Memory LibSQL + semantic recall
TypeScript-first agent framework. Default memory: LibSQL store + on-device fastembed-js for semantic recall.
- Microsoft Agent Governance Toolkit Cross-Model Verification Kernel + Agent OS
Cross-Model Verification Kernel (CMVK) requires majority-voting agreement across multiple model calls before a memory-influenced action. Agent OS package intercepts every action (memory included) at sub-millisecond latency. MIT licensed.
- Pydantic-AI + Hindsight External memory bolted on
Pydantic-AI ships no native persistent memory; agents start fresh each run. Long-term memory typically added via third-party integration (e.g. Hindsight) using tools + auto-injected memory instructions.
- Strands Agents Memory (AWS) Mem0 + AgentCore-backed
AWS code-first agent framework. Multi-LLM (Bedrock, Anthropic, Gemini, OpenAI). Memory via Mem0 integration (ElastiCache for Valkey + Neptune Analytics) or Bedrock AgentCore Memory. Local default uses FAISS; OpenSearch for production.
Tier 3 — emerging (1)
- Pydantic-AI MemoryTool Tool-delegated to model provider
Pydantic AI Harness ships a Memory capability: a MEMORY.md notebook plus extra Markdown notes, exposed via write/read/delete/search memory tools and auto-injected into context; backends InMemory, File, SQLite, Postgres with per-user name…