// SCAN REPORT

MCP Toolkit

28 tools. 17 can modify or destroy data without limits.

3 destructive tools with no built-in limits. Policy required.

Last updated: 30 Jun 2026

17 can modify or destroy data

11 read-only

28 tools total

Community server · catalogue entry verified 30/06/2026

How to control MCP Toolkit ↓

TOOL MAP

What MCP Toolkit exposes to your agents

Read (11) Write / Execute (14) Destructive / Financial (3)

Running more than one server? Check your whole stack →

MOST DANGEROUS TOOLS

Critical Risk

The most dangerous MCP Toolkit tools

Destructive · High delete_directory Deleting a directory is an irreversible destructive action that cannot be undone. Destructive · High delete_file This tool permanently removes files from the filesystem without the ability to undo the operation. Destructive · Critical github_delete_repo This tool irreversibly deletes an entire GitHub repository—a destructive operation that cannot be undone. Execute · Critical execute_command This tool runs system commands whose effects are entirely dependent on the arguments provided.

17 of MCP Toolkit's 28 tools can modify, destroy, or commit something on every call — and an agent calls them with no built-in limits.

POLICY

How to control MCP Toolkit

PolicyLayer is an MCP gateway — it sits between your AI agents and MCP Toolkit, and nothing reaches the server without passing your rules. These are the rules we recommend:

Deny destructive operations

{
  "delete_directory": {
    "deny_if": [
      {
        "conditions": [],
        "on_deny": "Blocked by default. Requires approval."
      }
    ]
  }
}

Destructive tools should never be available to autonomous agents without human approval.

Rate limit write operations

{
  "create_directory": {
    "limits": [
      {
        "counter": "create_directory_per_hour",
        "window": "hour",
        "max": 30,
        "scope": "grant"
      }
    ]
  }
}

Prevents bulk unintended modifications from agents caught in loops.

Cap read operations

{
  "brave_search": {
    "limits": [
      {
        "counter": "brave_search_per_minute",
        "window": "minute",
        "max": 60,
        "scope": "grant"
      }
    ]
  }
}

Controls API costs and prevents retry loops from exhausting upstream rate limits.

Create a free account and register MCP Toolkit — nothing to install.
Add these rules — paste them, or build them visually. Tune the limits to your setup.
Point your MCP client (Claude, Cursor, anything) at your gateway URL.

ENFORCE POLICY ON MCP TOOLKIT →

Instant setup, no code required.

FULL CATALOGUE

All 28 MCP Toolkit tools

DESTRUCTIVE 3 tools

Destructive delete_directory 删除目录 Destructive delete_file 删除文件 Destructive github_delete_repo 删除GitHub仓库。请谨慎使用此功能，删除后无法恢复！

EXECUTE 5 tools

Execute db_begin_transaction 开始数据库事务。当我需要确保多个操作的原子性时使用。仅支持 MySQL 和 PostgreSQL。 Execute db_connect 连接到数据库。在执行任何数据库操作前，我会先使用此工具建立连接。我会自动从配置文件读取连接信息，无需手动输入敏感信息。 Execute db_query 执行数据库查询。我会使用参数化查询来防止 SQL 注入。对于 Redis，我会自动将命令解析为正确的格式。 Execute execute_command 执行命令行命令 Execute http_request 发送HTTP请求并返回完整响应信息

WRITE 9 tools

Write create_directory 创建目录 Write db_close 关闭数据库连接。我会在完成操作后调用此工具释放资源。如果有活动事务会自动回滚。 Write db_commit_transaction 提交数据库事务。确认事务中的所有更改。如果事务中有任何操作失败，我会使用 db_rollback_transaction。 Write db_rollback_transaction 回滚数据库事务。当发生错误或需要撤销更改时，我会使用此工具来取消事务中的所有更改。 Write github_create_repo 创建一个新的GitHub仓库。 Write github_update_repo 更新GitHub仓库的设置。 Write logger 写入日志到当前工作目录的log目录下,按日期分文件存储 Write replace_in_file 替换文件内容（支持多块替换） Write write_file 写入文件

READ 11 tools

Read brave_search 使用Brave搜索引擎搜索网页。当需要搜索网页内容时优先使用此工具,而不是使用浏览器。支持高级搜索语法,结果更准确且无广告。 Read everything_search 使用 Everything 搜索本地文件。当需要在本地快速查找文件时，优先使用此工具而不是操作系统的搜索功能。支持正则表达式和高级搜索语法，搜索速度极快。 Read get_stack_trace 从错误输出中提取堆栈跟踪信息 Read github_cat 查看 GitHub 仓库中的文件内容。当我需要分析或参考特定文件的实现时使用此工具。 Read github_list_repos 列出当前用户的所有仓库。用于查看和管理你的GitHub仓库。 Read github_ls 列出 GitHub 仓库中的文件和目录。当需要浏览 GitHub 仓库内容时使用此工具。我会自动从配置文件读取 token 和代理设置。 Read github_search_code 在 GitHub 上搜索代码。当我需要查找特定实现方式、解决方案或代码示例时，我会使用此工具。我会根据上下文自动构建精确的搜索关键词。 Read github_search_repo 搜索 GitHub 仓库。当我需要查找特定主题、功能或示例代码的仓库时，我会使用此工具。我会根据上下文自动构建合适的搜索关键词。 Read github_tree 递归显示 GitHub 仓库的完整目录树结构。当需要了解仓库的整体文件组织时使用此工具，它提供比 github_ls 更全面的视图。 Read read_file 读取文件 Read read_webpage 读取网页的主要内容。当需要阅读特定网页的文本内容时使用此工具（不包括 GitHub，GitHub 相关内容请使用专门的 GitHub 工具）。此工具会自动提取主要内容，忽略广告、导航栏等干扰元素。支持普通 HTML 页面

EXPLORE

Related servers

Other MCP servers with similar tools — same risk classification, starter policies for each.

FAQ

Questions about MCP Toolkit

Can an AI agent delete data through the MCP Toolkit MCP server? +

Yes. The MCP Toolkit server exposes 3 destructive tools including delete_directory, delete_file, github_delete_repo. These permanently remove resources with no undo. PolicyLayer blocks destructive tools by default so they never reach the upstream server.

How do I prevent bulk modifications through MCP Toolkit? +

The MCP Toolkit server has 9 write tools including create_directory, db_close, db_commit_transaction. Set a rate limit in your policy -- for example, 10 calls per hour prevents an agent from making more than 10 modifications per hour. PolicyLayer enforces this at the gateway, before calls reach MCP Toolkit.

How many tools does the MCP Toolkit MCP server expose? +

28 tools across 4 categories: Destructive, Execute, Read, Write. 11 are read-only. 17 can modify, create, or delete data.

How do I enforce a policy on MCP Toolkit? +

Register the MCP Toolkit MCP server in PolicyLayer, apply the suggested rules above (adjust the limits to your use case), and point your AI client at the PolicyLayer proxy URL instead of the server directly. Your agents keep the same tools; PolicyLayer evaluates every call against policy before it executes. Nothing to install, live in minutes.

Enforce policy on every MCP Toolkit tool call.

Deterministic rules across all 28 MCP Toolkit tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.

ENFORCE POLICY ON MCP TOOLKIT →

Instant setup, no code required.

28 MCP Toolkit tools catalogued and risk-classified — across an index of 43,000+ MCP servers.