加载中…
Persistent memory MCP server. 28 tools, BM25+MMR+OOD retrieval, tamper-evident audit, code graph.
io.github.LeandroPG19/cuba-memorys 是 ChatHome Skills 收录的 MCP 条目，来源为 Official MCP Registry，质量分 73/100。 作为 Model Context Protocol 服务器，它向 Claude、Cursor、Copilot 等兼容 MCP 的 LLM 客户端暴露可直接调用的工具，让 agent 的能力超出模型内置范围。 该条目由 ChatHome Skills 的每日抓取从官方 registry 和 GitHub 发现查询中收录，每次运行都会重新核验仓库事实、新鲜度信号和安装元数据。 质量分综合了官方 registry 收录情况、可用安装命令、许可证完整性、仓库活跃度和 GitHub star 趋势，完整算法说明见来源页。安装命令已列于上方，可直接复制使用。 在将 io.github.LeandroPG19/cuba-memorys 加入生产 agent 工作流前，请核对上方的最近更新日期和仓库状态——下方事实快照是最近一次抓取的时点值。
npx -y cuba-memorys@0.24.0运行此命令前，请先核对上游源码。
{
"_meta": {
"io.modelcontextprotocol.registry/official": {
"status": "active",
"isLatest": true,
"updatedAt": "2026-08-16T19:27:16.321527Z",
"publishedAt": "2026-08-16T19:27:16.321527Z",
"statusChangedAt": "2026-08-16T19:27:16.321527Z"
}
},
"server": {
"name": "io.github.LeandroPG19/cuba-memorys",
"$schema": "https://static.modelcontextprotocol.io/schemas/2025-12-11/server.schema.json",
"version": "0.24.0",
"packages": [
{
"version": "0.24.0",
"transport": {
"type": "stdio"
},
"identifier": "cuba-memorys",
"registryType": "npm",
"environmentVariables": [
{
"name": "DATABASE_URL",
"format": "string",
"isRequired": true,
"description": "PostgreSQL connection URL (e.g. postgresql://user:pass@localhost:5432/brain)"
},
{
"name": "CUBA_PROJECT_FILTER",
"format": "string",
"description": "Set to 'off' to disable v0.8 project scoping filters (admin/debug only)."
},
{
"name": "CUBA_SYNC_DIR",
"format": "string",
"description": "Directory used by cuba_sync export/import (default ./.cuba-memorys/)."
},
{
"name": "CUBA_JUDGE",
"format": "string",
"description": "cuba_juez backend selector: claude_cli | anthropic_api | heuristic | auto (default auto)."
},
{
"name": "CUBA_JUEZ_CLI",
"format": "string",
"description": "Subprocess CLI used by ClaudeCodeJudge (default 'claude')."
},
{
"name": "CUBA_JUEZ_MODEL",
"format": "string",
"description": "Model name passed to the judge backend (default 'claude-haiku-4-5')."
},
{
"name": "ANTHROPIC_API_KEY",
"format": "string",
"isSecret": true,
"description": "API key for AnthropicApiJudge (only used when feature 'anthropic-api' is built in)."
},
{
"name": "CUBA_REM_RELATION_BATCH",
"format": "string",
"description": "Isolated entities the REM cycle scans for relations per run (default 5, 0 disables)."
},
{
"name": "CUBA_RERANKER_PATH",
"format": "string",
"description": "Directory holding the bge-reranker ONNX model; enables cross-encoder reranking (+93% nDCG on GPU)."
},
{
"name": "CUBA_RERANK_FIXED_SHAPE",
"format": "string",
"description": "Pad reranker batches to a constant 512 tokens. Defaults on for GPU builds, off on CPU."
}
]
},
{
"version": "1.26.0",
"transport": {
"type": "stdio"
},
"identifier": "cuba-memorys",
"registryType": "pypi",
"environmentVariables": [
{
"name": "DATABASE_URL",
"format": "string",
"isRequired": true,
"description": "PostgreSQL connection URL (e.g. postgresql://user:pass@localhost:5432/brain)"
}
]
}
],
"repository": {
"url": "https://github.com/LeandroPG19/cuba-memorys",
"source": "github"
},
"description": "Persistent memory MCP server. 28 tools, BM25+MMR+OOD retrieval, tamper-evident audit, code graph."
}
}