MCP低风险未认领
cogvault
Local hybrid-search memory for AI agents over plain Markdown. Multi-tenant, offline, no LLM.
NBibikovnbibikov/cogvault
server.json
{
"$schema": "https://static.modelcontextprotocol.io/schemas/2025-12-11/server.schema.json",
"name": "io.github.NBibikov/cogvault",
"description": "Local hybrid-search memory for AI agents over plain Markdown. Multi-tenant, offline, no LLM.",
"title": "cogvault",
"repository": {
"url": "https://github.com/NBibikov/cogvault",
"source": "github"
},
"version": "0.11.3",
"websiteUrl": "https://github.com/NBibikov/cogvault",
"icons": [
{
"src": "https://raw.githubusercontent.com/NBibikov/cogvault/main/assets/logo.png",
"mimeType": "image/png",
"sizes": [
"512x512"
]
}
],
"packages": [
{
"registryType": "pypi",
"identifier": "cogvault",
"version": "0.11.3",
"runtimeHint": "uvx",
"transport": {
"type": "stdio"
},
"packageArguments": [
{
"value": "mcp",
"type": "positional",
"valueHint": "command"
},
{
"description": "Directory of Markdown memory cards for this agent (one directory = one isolated tenant)",
"isRequired": true,
"format": "filepath",
"placeholder": "~/agent/memory",
"type": "named",
"name": "--tenant"
}
],
"environmentVariables": [
{
"description": "Embedding model override; otherwise .cogvault.toml in the tenant, then the multilingual default",
"format": "string",
"name": "COGVAULT_MODEL"
},
{
"description": "Recall log: 'off' to disable, or a path to relocate the JSONL log",
"format": "string",
"name": "COGVAULT_LOG"
}
]
}
]
}权限
声明检测
运行代码—
python安装—
pypi:cogvault@0.11.3安装时运行脚本—无
网络无无
需要的凭据无无
工作区外的路径—无
智能体工具—无
检查
低风险 · 没有发现需要提醒的地方。
未经人工审核 · 已做规则检查;模型审核尚未开启。
版本
- #10.11.3最新2026年10月7日
cogvault在 Codeg 中打开