// SCAN REPORT

Mcp Swarm

42 tools. 18 can modify or destroy data without limits.

1 destructive tool with no built-in limits. Policy required.

Last updated: 3 Jul 2026

18 can modify or destroy data

24 read-only

42 tools total

Community server · catalogue entry verified 03/07/2026

How to control Mcp Swarm ↓

TOOL MAP

What Mcp Swarm exposes to your agents

Read (24) Write / Execute (17) Destructive / Financial (1)

Running more than one server? Check your whole stack →

MOST DANGEROUS TOOLS

Critical Risk

The most dangerous Mcp Swarm tools

Destructive · High swarm_reset_circuit_breaker 重置指定模型的熔断器状态，使其恢复正常服务。 Execute · Medium swarm_build_index Building an index requires scanning the entire codebase and writing index data to storage. Execute · High swarm_execute_skill This tool triggers external operations (code review, document generation, and other skills) whose side effects depend entirely on which skill is selected and its arguments. Execute · High swarm_execute_swarm This tool executes pre-defined agent swarm tasks using various coordination strategies.

18 of Mcp Swarm's 42 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 Swarm

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

Deny destructive operations

{
  "swarm_reset_circuit_breaker": {
    "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

{
  "swarm_autofix_lint": {
    "limits": [
      {
        "counter": "swarm_autofix_lint_per_hour",
        "window": "hour",
        "max": 30,
        "scope": "grant"
      }
    ]
  }
}

Prevents bulk unintended modifications from agents caught in loops.

Cap read operations

{
  "swarm_analyze_task": {
    "limits": [
      {
        "counter": "swarm_analyze_task_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 Swarm — 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 SWARM →

Instant setup, no code required.

FULL CATALOGUE

All 42 Mcp Swarm tools

DESTRUCTIVE 1 tools

Destructive swarm_reset_circuit_breaker 重置指定模型的熔断器状态，使其恢复正常服务。

EXECUTE 10 tools

Execute swarm_build_index 构建整个代码库的索引，用于快速搜索和导航。 Execute swarm_chat 根据任务内容自动选择最合适的模型并调用你的 API 进行对话。支持 8 个模型：qwen3.5-plus、qwen3-max-2026-01-23、qwen3-coder-next、qwen3-coder-plus、Mi Execute swarm_chat_with_model 使用指定模型调用 API。可选模型： Execute swarm_create_dynamic_agents 根据任务描述自动生成多个 specialized sub-agents，实现自组织结构（参考 Kimi K2.5 Agent Swarm）。 Execute swarm_execute_skill 执行指定的技能（如代码审查、文档生成等）。 Execute swarm_execute_swarm 执行已创建的 Agent 蜂群任务，支持并行、辩论、Map-Reduce 等协调策略。 Execute swarm_execute_workflow 执行一个多步骤的工作流。支持内置工作流或自定义工作流定义。 Execute swarm_run_lint 运行代码检查工具。 Execute swarm_run_test 运行测试套件。 Execute swarm_validate_with_gates 使用质量门控验证内容（代码、文档等）。

WRITE 7 tools

Write swarm_autofix_lint 自动修复代码检查错误。 Write swarm_generate_mr_description 生成 Merge Request / Pull Request 描述模板。 Write swarm_git_commit 提交代码变更。 Write swarm_git_create_branch 创建新的 Git 分支。 Write swarm_modify_file 智能修改文件内容（查找替换）。会先备份现有文件。 Write swarm_vector_upsert 将一段文本向量化并写入向量库，便于后续向量搜索。 Write swarm_write_file 创建或覆盖文件。会先备份现有文件。

READ 24 tools

Read swarm_analyze_task 根据任务描述分析任务类型并返回推荐模型与备选列表，不调用 API。 Read swarm_analyze_workspace 扫描工作空间所有项目，分析架构、技术栈和问题。 Read swarm_cache_stats 获取 Redis 缓存统计信息。 Read swarm_code_stats 获取 MongoDB 代码存储的统计信息。 Read swarm_db_stats 获取数据库连接和存储统计信息。 Read swarm_detect_tech_stack 自动识别项目使用的技术栈，包括框架、库、工具等。 Read swarm_generate_agent_roles 仅生成 Agent 角色定义（不执行），用于查看系统会创建哪些 specialized sub-agents。 Read swarm_get_lock_stats 获取文件锁的统计信息。 Read swarm_get_logs 获取最近的系统日志，支持按级别过滤。 Read swarm_get_metrics 获取蜂群系统的性能指标，包括响应时间、成功率等。 Read swarm_get_stats 获取蜂群系统的使用统计信息，包括请求次数、模型使用分布、成本监控、错误统计等。 Read swarm_git_status 获取 Git 仓库的当前状态。 Read swarm_list_agents 列出所有可用的专业 Agent，支持按类别筛选。 Read swarm_list_directory 列出项目目录结构。 Read swarm_list_gates 列出所有可用的质量门控。 Read swarm_list_skills 列出所有可用的技能。 Read swarm_list_workflows 列出所有可用的内置工作流模板。 Read swarm_read_file 读取指定文件的内容。需要提供文件路径。 Read swarm_recommend_agent 根据任务描述智能推荐最合适的专业 Agent。 Read swarm_recommend_workflow 根据任务描述智能推荐最合适的工作流。 Read swarm_search_code 在代码库中搜索文本内容（使用 ripgrep）。 Read swarm_search_files 按文件名模式搜索文件。 Read swarm_validate_workflow 验证自定义工作流定义是否有效。 Read swarm_vector_search 使用向量相似度搜索代码或文档。

EXPLORE

Related servers

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

FAQ

Questions about Mcp Swarm

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

Yes. The Mcp Swarm server exposes 1 destructive tools including swarm_reset_circuit_breaker. 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 Swarm? +

The Mcp Swarm server has 7 write tools including swarm_autofix_lint, swarm_generate_mr_description, swarm_git_commit. 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 Swarm.

How many tools does the Mcp Swarm MCP server expose? +

42 tools across 4 categories: Destructive, Execute, Read, Write. 24 are read-only. 18 can modify, create, or delete data.

How do I enforce a policy on Mcp Swarm? +

Register the Mcp Swarm 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 Swarm tool call.

Deterministic rules across all 42 Mcp Swarm tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.

ENFORCE POLICY ON MCP SWARM →

Instant setup, no code required.

42 Mcp Swarm tools catalogued and risk-classified — across an index of 43,000+ MCP servers.