// SCAN REPORT

Prompts

23 tools. 12 can modify or destroy data without limits.

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

Last updated: 3 Jul 2026

12 can modify or destroy data

11 read-only

23 tools total

Community server · catalogue entry verified 03/07/2026

How to control Prompts ↓

TOOL MAP

What Prompts exposes to your agents

Read (11) Write / Execute (11) Destructive / Financial (1)

Running more than one server? Check your whole stack →

MOST DANGEROUS TOOLS

Critical Risk

The most dangerous Prompts tools

Destructive · Medium remove_rule This tool permanently removes project governance rules from the system. Execute · Medium commit_dialog This tool executes a git commit operation, which is an external VCS action that modifies repository history. Execute · Critical safe_publish This tool executes a multi-step pipeline: it publishes to npm (external package registry), pushes to git (modifies remote repository), bumps version numbers, updates README, and runs npm link. Write · Medium add_rule This tool creates new rules that are stored persistently in a project system.

12 of Prompts's 23 tools can modify, destroy, or commit something on every call — and an agent calls them with no built-in limits.

POLICY

How to control Prompts

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

Deny destructive operations

{
  "remove_rule": {
    "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

{
  "add_rule": {
    "limits": [
      {
        "counter": "add_rule_per_hour",
        "window": "hour",
        "max": 30,
        "scope": "grant"
      }
    ]
  }
}

Prevents bulk unintended modifications from agents caught in loops.

Cap read operations

{
  "audit_secrets": {
    "limits": [
      {
        "counter": "audit_secrets_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 Prompts — 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 PROMPTS →

Instant setup, no code required.

FULL CATALOGUE

All 23 Prompts tools

DESTRUCTIVE 1 tools

Destructive remove_rule 【删除规则】删除一条项目规范规则。

EXECUTE 2 tools

Execute commit_dialog 【手动提交】手动触发一次 git commit。可指定要提交的文件，不指定则提交所有变更。 Execute safe_publish 【安全发布】一键完成隐私审计 + 版本 bump + npm publish + git push + npm link + README 更新。

WRITE 9 tools

Write add_rule 【添加规则】添加一条项目规范规则。规则会持久化存储，在每次会话启动时自动加载。 Write add_skill 【创建技能】创建一个新的角色技能。可自定义身份、开发规范等。 Write confirm_direction 【方向确认】AI 追问澄清后，用户确认方向时调用。保存目标、约束、验收标准到 direction.md。对话式追问由 AI 自然完成，此工具只负责持久化确认结果。 Write init_prompts 【初始化】扫描目标项目，自动生成 prompts 体系（context.md / todos.md / modules/）。已有文件不会覆盖。 Write log_dialog 【记录日志】记录对话日志，追加待办到 todos.md。 Write log_module 【模块记录】按模块记录一次修改（目录式）。修改功能前先 read_module，修改后调用此工具。 Write onboard_tool 【工具引导】为指定工具生成 3 步快速上手引导（是什么 → 用一次 → 验证效果）。 Write update_skill 【技能自优化】会话结束时调用，总结本次开发经验并更新 Skill。可追加学习记录、修改开发规范、更新描述。智能体应主动在每次开发后调用此工具自我进化。 Write update_todos 【更新待办】更新 todos.md 中的待办事项。

READ 11 tools

Read audit_secrets 【隐私审计】扫描代码中的密钥、Token、密码等隐私信息。发布前必须调用，有 critical/high 发现时拦截发布。 Read auto_start 【自动启动】会话开始时第一个调用。一键加载全部上下文（context + todos + 用户规则 + 模块记录）。每次新对话开始时必须调用此工具。 Read bootstrap 【一键启动】自动读取项目上下文（context.md + todos + 模块记录）。智能体启动时第一步调用。 Read list_rules 【列出规则】列出所有已添加的项目规范规则。 Read list_scenes 【场景列表】列出所有可用的开发场景及推荐工具。 Read list_skills 【技能列表】列出所有可用的角色技能（Skill）。会话启动时自动展示，也可手动调用。 Read onboard_all 【全部引导】为所有已安装工具生成引导手册概览。 Read read_module 【读取模块记录】修改功能前调用，读取对应模块的历史修改记录。 Read recommend_tools 【场景推荐】根据当前开发场景（coding/testing/debugging/reviewing 等），推荐应使用的工具和操作建议。也可自动检测场景。 Read scan_tools 【工具扫描】扫描项目已安装的工具（npm 依赖、MCP Server），输出能力清单和使用建议。帮助用户快速了解可用工具。 Read select_skill 【选择技能】选择一个 Skill 作为当前身份角色。返回该 Skill 的完整 prompt（身份 + 开发规范 + 学习记录）。会话开始时应询问用户选择哪个 Skill。

EXPLORE

Related servers

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

FAQ

Questions about Prompts

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

Yes. The Prompts server exposes 1 destructive tools including remove_rule. 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 Prompts? +

The Prompts server has 9 write tools including add_rule, add_skill, confirm_direction. 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 Prompts.

How many tools does the Prompts MCP server expose? +

23 tools across 4 categories: Destructive, Execute, Read, Write. 11 are read-only. 12 can modify, create, or delete data.

How do I enforce a policy on Prompts? +

Register the Prompts 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 Prompts tool call.

Deterministic rules across all 23 Prompts tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.

ENFORCE POLICY ON PROMPTS →

Instant setup, no code required.

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