Agent Permissions Index
Grade Claude Code, Gemini CLI and Cursor configs for risky permissions; get public config stats.
我该使用它吗
质量与安全性
B
基于对工具定义和协议合规性的自动分析。
上下文开销
~309token 数（工具定义）
~611 B典型响应大小
对注意力的影响极小（占 128k 上下文窗口的 0.24%）
这是每次将服务器的工具加载到模型上下文窗口时所消耗的大致 token 数。数值越高，可用于其他任务的注意力就越少。
安装
一键安装
将以下内容添加到你的 `claude_desktop_config.json` 文件中：
{
"mcpServers": {
"agent-permissions-index": {
"url": "https://agent-permissions-index.vercel.app/mcp"
}
}
}远程端点
https://agent-permissions-index.vercel.app/mcpstreamable-http它能做什么
工具清单
工具（3）
🟢 只读🟡 写入🔴 删除⚪ 未知
⚪grade_config(config, tool)
Grade an AI coding agent permissions config (Claude Code .claude/settings.json, Gemini CLI .gemini/settings.json, mcp.json, or Codex CLI config.toml). Returns a letter grade, findings, a safer config and README badge Markdown. Inputs are not stored or logged.
输入模式
{
"type": "object",
"properties": {
"config": {
"type": "string",
"minLength": 1,
"maxLength": 100000,
"description": "The config file contents as JSON text, or TOML for Codex config.toml (max 100 KB)"
},
"tool": {
"default": "claude_code",
"description": "Which tool the config is for",
"type": "string",
"enum": [
"claude_code",
"gemini_cli",
"mcp_json",
"cursor_mcp",
"codex"
]
}
},
"required": [
"config"
],
"$schema": "https://json-schema.org/draft/2020-12/schema"
}🟢get_index_stats
Headline aggregate numbers from the latest weekly snapshot of public AI agent configs on GitHub.
输入模式
{
"type": "object",
"properties": {},
"$schema": "https://json-schema.org/draft/2020-12/schema"
}🟡get_leaderboard
Agent tools ranked by how safely their public configs are set up, with sample sizes and method.
输入模式
{
"type": "object",
"properties": {},
"$schema": "https://json-schema.org/draft/2020-12/schema"
}社区
证据
最近观测
已验证未记录版本3 个工具