Skip to main content
Glama
promptz.dev MCP Server

promptz.dev MCP Server

by cremich

promptz.dev MCP 服务器

在 Amazon Q Developer 中直接访问来自 promptz.dev 的提示。

此 MCP 服务器允许从 promptz.dev API 访问提示而无需复制粘贴,从而减少开发工作流程中的上下文切换和摩擦。

特征

promptz.dev MCP 服务器提供两项主要功能:

  1. 提示——可执行函数来搜索和执行提示。

  2. 规则- 可执行函数用于搜索项目规则并通过与其他工具集成在您的工作区中添加/更新它们。

Related MCP server: prompts.chat MCP Server

示例用法

一旦服务器连接到 Amazon Q Developer,您就可以使用自然语言,例如:

  • “搜索有关 JavaScript 的 CLI 提示”

  • “向我显示名为‘React 组件文档’的提示”

  • “使用 React 组件文档提示来改进我的文档”

  • “查找 CDK 开发的项目规则”

  • “将 CDK 项目结构项目规则添加到我的工作区”

安装

步骤 1:获取 API 凭证

  1. 导航至https://promptz.dev/mcp

  2. 复制 MCP 设置,如 API 密钥、API URL 或示例 MCP 配置片段。

步骤 2:安装 MCP 服务器

打开位于~/.aws/amazonq/mcp.json的 Amazon Q Developer MCP 客户端设置文件

选项 1:使用 npx(推荐)

使用服务器的最简单方法是使用 npx,它不需要安装:

  1. 将以下配置添加到您的 Amazon Q Developer MCP 客户端的设置文件中:

{
  "mcpServers": {
    "promptz.dev": {
      "command": "npx",
      "args": ["-y", "@promptz/mcp"],
      "env": {
        "PROMPTZ_API_URL": "your-api-url-from-promptz.dev",
        "PROMPTZ_API_KEY": "your-api-key-from-promptz.dev"
      },
      "disabled": false,
      "autoApprove": []
    }
  }
}

选项 2:本地安装

  1. 克隆存储库:

git clone https://github.com/cremich/promptz-mcp.git
cd promptz-mcp
  1. 安装依赖项并构建:

npm install
npm run build
  1. 将以下配置添加到您的 MCP 客户端的设置文件:

{
  "mcpServers": {
    "promptz.dev": {
      "command": "node",
      "args": ["/path/to/promptz-mcp/build/index.js"],
      "env": {
        "PROMPTZ_API_URL": "your-api-url-from-promptz.dev",
        "PROMPTZ_API_KEY": "your-api-key-from-promptz.dev"
      },
      "disabled": false,
      "autoApprove": []
    }
  }
}

故障排除

如果您遇到服务器问题:

  1. 检查您的 API 凭证是否正确

  2. 确保服务器在 MCP 客户端中正确配置

  3. 在位于~/.promptz/logs/mcp-server.log日志中查找错误消息

  4. 使用 MCP Inspector 进行调试:

# Run with environment variables
PROMPTZ_API_URL="your-api-url" PROMPTZ_API_KEY="your-api-key" npm run inspector

检查器将提供一个 URL 来访问浏览器中的调试工具。

发展

对于那些想要贡献或修改服务器的人:

# Install dependencies
npm install

# Build the server
npm run build

# For development with auto-rebuild
npm run watch

# Run tests
npm test

安全注意事项

  • 该服务器仅提供对提示的读取权限,不执行任何写入操作

  • API 凭证存储在您的 MCP 客户端的配置文件中

  • 与 promptz.dev API 的所有通信均通过 HTTPS 完成

  • 服务器记录到您的主目录中的文件(~/.promptz/logs/mcp-server.log)

Available Tools

4 tools
get_promptC

Get a specific prompt by ID or name

Inputs: name.

get_ruleC

Get a specific project rule by name

Inputs: name.

list_promptsC

List available prompts from promptz.dev

Inputs: cursor, tags.

list_rulesB

List available project rules from promptz.dev

Inputs: cursor, tags.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    B3.4/5.0

    Scored across 4 tools

    Disambiguation5/5

    Each tool has a clearly distinct purpose: get_prompt retrieves a specific prompt, get_rule retrieves a specific rule, list_prompts lists all prompts, and list_rules lists all rules. There is no overlap or ambiguity between these four operations.

    Naming Consistency5/5

    All tool names follow a consistent verb_noun pattern with snake_case (get_prompt, get_rule, list_prompts, list_rules). The naming is perfectly predictable and readable throughout the set.

    Tool Count4/5

    With 4 tools, the count is reasonable and well-scoped for a prompt management server, covering core retrieval and listing operations. It's slightly lean but appropriate for the apparent scope, lacking only minor gaps like update or delete operations.

    Completeness3/5

    The tool set provides good read/list coverage for prompts and rules, but there are notable gaps in CRUD operations. For example, there are no tools for creating, updating, or deleting prompts or rules, which limits full lifecycle management for the domain.

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      B
      quality
      D
      maintenance
      Provides access to thousands of AI prompts from prompts.chat, enabling users to search, browse, and use community-curated prompts with variable substitution directly in their AI coding assistant.
      2
      339 npm
      40
      ISC
    • A
      license
      Not graded
      quality
      F
      maintenance
      Connects AI tools to your organization's Haiven prompts, allowing seamless access to expert-crafted prompts for user stories, code reviews, and more without switching apps.
      Apache 2.0