GoodMem

• CollectionMemory & State
• Classificationagent-native
• MCPAvailable
• Agent SkillAvailable
• InterfacesAgent Skill · MCP · REST · CLI · SDK · STDIO · http · API

• Website ↗
• Docs ↗
• Official repo ↗
• Full dossier on GitHub ↗

What it does

GoodMem gives agents persistent, searchable memory outside their context windows. Agents store documents, prior findings, and task history in memory spaces. The service handles configured chunking and embedding, then retrieves relevant context with optional reranking. Spaces can be isolated for one workload or shared between authorized agents. See the integration overview.

Dual-use boundary: GoodMem also supports conventional RAG, including document Q&A and knowledge search. Its storage and retrieval APIs are useful to ordinary applications as well as agents; those APIs are not exclusively agent-specific. This entry covers the persistent agent-memory workflow, published MCP/skills, and workload identity model. It does not claim automatic conversation-fact extraction, conflict resolution, or autonomous forgetting merely from storing a memory.

Agent-facing interfaces

• Interaction pattern: MCP (Streamable HTTP)
• Catalog protocols: Agent Skill, MCP, REST, CLI, SDK, STDIO, http, API
• MCP: ✅ Available through two distinct interfaces.
• Agent Skill: ✅ Available in the official Claude Code plugin.
• Onboarding type: mcp

How agents onboard

Interaction pattern: MCP (Streamable HTTP)

{
  "mcpServers": {
    "goodmem": {
      "url": "https://goodmem.example.com/mcp",
      "headers": {
        "x-api-key": "gm_your_key"
      }
    }
  }
}

See the full onboarding section in the dossier.

Protocol surface