ai-runtime-guard
ai-runtime-guard is an MCP server that sits between your AI agent and your system, enforcing a policy layer before any file or shell action takes effect. No retraining, no prompt engineering, no changes to your agent or workflow, just install, configure once, and your agent operates within the boundaries you set.
安全与权限
0 个平台
0 项功能
安装
uvx ai-runtime-guard
把上面的配置写入 MCP 客户端的配置文件（Claude Desktop 为 claude_desktop_config.json），重启客户端后生效。
{
"mcpServers": {
"runtime-guard": {
"command": "uvx",
"args": [
"ai-runtime-guard"
],
"env": {
"AIRG_AGENT_ID": "<AIRG_AGENT_ID>",
"AIRG_LOG_PATH": "<AIRG_LOG_PATH>",
"AIRG_WORKSPACE": "<AIRG_WORKSPACE>",
"AIRG_POLICY_PATH": "<AIRG_POLICY_PATH>",
"AIRG_SERVER_COMMAND": "<AIRG_SERVER_COMMAND>",
"AIRG_REPORTS_DB_PATH": "<AIRG_REPORTS_DB_PATH>",
"AIRG_APPROVAL_DB_PATH": "<AIRG_APPROVAL_DB_PATH>",
"AIRG_APPROVAL_HMAC_KEY_PATH": "<AIRG_APPROVAL_HMAC_KEY_PATH>"
}
}
}
}把上面的配置写入 MCP 客户端的配置文件（Claude Desktop 为 claude_desktop_config.json），重启客户端后生效。
数据来源：公开的 MCP Server 目录。本站为独立第三方目录，与各 MCP Server 维护方均无隶属或背书关系；名称与简介如实标注，介绍文案由本站再加工。
