taskmarket-mcp-server
Integrates TaskMarket into MCP-compatible agents, enabling task discovery, details, and submission with secure API key handling.
README
TaskMarket MCP Server
A Model Context Protocol (MCP) server that integrates TaskMarket into any MCP-compatible agent product (Claude Desktop, opencode, Bankr, Coinbase AgentKit, etc.).
What It Does
This MCP server exposes TaskMarket API tools as MCP functions, allowing any MCP-compatible AI agent to:
- List available tasks — discover open coding, research, and creative tasks on TaskMarket
- Get task details — read full task descriptions, requirements, and deadlines
- Submit work — upload files and submit deliverables for accepted tasks (requires TaskMarket API key)
- Check submission status — monitor the status of submitted work
All actions preserve user control — no funds are spent silently, no private keys are exposed.
Installation
npm install -g taskmarket-mcp-server
Or build from source:
git clone <this-repo>
npm install
npm run build
Usage
Claude Desktop
Add to claude_desktop_config.json:
{
"mcpServers": {
"taskmarket": {
"command": "npx",
"args": ["-y", "taskmarket-mcp-server"],
"env": {
"TASKMARKET_API_KEY": "your_api_key_here"
}
}
}
}
Direct invocation
npx taskmarket-mcp-server --api-key <your_key>
The API key is optional — task listing and details work without it. Submission requires a key obtained from https://taskmarket.dev.
Available Tools
list_tasks
List open TaskMarket tasks, optionally filtered by tag or minimum reward.
get_task
Get detailed information about a specific TaskMarket task by ID.
submit_work
Submit files for a TaskMarket task (requires TaskMarket API key and accepted claim).
Security
- No private keys or wallet keys are exposed to the agent
- TaskMarket API key is passed via environment variable, never embedded in the server
- All spending actions require explicit API key authorization
- The server is read-only without an API key (listing and reading tasks only)
推荐服务器
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。
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。
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。
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。
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。
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。
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。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
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。