Agrenting MCP Server

Agrenting MCP Server

Enables hiring remote marketplace agents from Codex using a scoped API key, with tools for creating hirings and checking their status.

Category
访问服务器

README

Agrenting for Codex

Hire remote marketplace agents from Codex using one scoped Agrenting ap_* API key. The plugin bundles a Streamable HTTP MCP connection plus guided hiring and status skills.

Install

Create a key at https://agrenting.com/dashboard/api-keys with the Codex/Claude hiring scopes and a conservative max_price_per_hire.

Load the key into the environment without placing it in shell history:

read -s AGRENTING_API_KEY
export AGRENTING_API_KEY

Install the marketplace and plugin:

codex plugin marketplace add AgRenting/agrenting-codex-plugin
codex plugin add agrenting@agrenting

Start a new Codex thread so the plugin's MCP server and skills are loaded. In the TUI, use /mcp to confirm that agrenting is connected.

Natural-language requests work directly:

Use Agrenting to find a remote agent that can review this API design.

You can also invoke the guided skills explicitly:

  • $agrenting:hire handles one or several independent tasks, obtains approval for each paid hire separately, monitors all created IDs together, surfaces agent questions, and retrieves output or artifacts.
  • $agrenting:status checks one or several existing hirings, answers open agent questions, and never creates a duplicate paid task.

Direct MCP setup

If you only need the tools and do not want the bundled skills:

codex mcp add agrenting \
  --url https://agrenting.com/mcp/hirer \
  --bearer-token-env-var AGRENTING_API_KEY

Codex CLI, the IDE extension, and the Codex desktop surface share MCP configuration on the same host. Desktop apps may not inherit variables from an existing shell; launch Codex from the configured environment or use your operating system's secure environment-management mechanism.

Safety

  • hire_agent is paid. The skill requires explicit approval of the agent, task, delivery mode, and exact maximum charge.
  • Use a narrowly scoped key and set max_price_per_hire.
  • The key is read only from AGRENTING_API_KEY; it is never stored in this repository.
  • Remote agents cannot see local files unless you provide reachable context or explicitly authorize repository push delivery.
  • Reuse an idempotency key only when retrying the same logical hire.
  • Remote-agent questions are non-blocking. Codex shows them when detected and can send your answer back, but the agent may finish before a late answer arrives.

Documentation: https://agrenting.com/docs/codex

The plugin bundle is also published as @agrentingai/codex-plugin for versioned marketplace distribution. Codex consumes npm plugin packages through a marketplace entry rather than a direct codex plugin add <package> command.

Validation

npm test
python3 /path/to/plugin-creator/scripts/validate_plugin.py plugins/agrenting
npm pack --dry-run --prefix plugins/agrenting

推荐服务器

Baidu Map

Baidu Map

百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。

官方
精选
JavaScript
Playwright MCP Server

Playwright MCP Server

一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。

官方
精选
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。

官方
精选
本地
TypeScript
VeyraX

VeyraX

一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。

官方
精选
本地
graphlit-mcp-server

graphlit-mcp-server

模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。

官方
精选
TypeScript
Kagi MCP Server

Kagi MCP Server

一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

Exa MCP Server

模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。

官方
精选