upwork-mcp

upwork-mcp

Connects AI agents to Upwork's GraphQL API, enabling job discovery, proposal management, profile tracking, and analytics.

Category
访问服务器

README

Upwork MCP Server

npm version CI License: MIT codecov

A Model Context Protocol server that connects AI agents to Upwork's GraphQL API — enabling job discovery, proposal management, profile tracking, and analytics.

Features

  • Official API Integration: Uses Upwork's GraphQL API with OAuth 2.0 authentication
  • Job Search: Search and filter job postings with advanced criteria
  • Proposal Management: Submit, update, track, and withdraw proposals
  • Saved Jobs: Bookmark jobs for later review and get personalized recommendations
  • Profile Management: View your freelancer profile, skills, and completeness
  • Contract Tracking: Monitor active contracts and earnings
  • Time Reports: Query work history and time logs
  • Rate Limiting: Built-in rate limiting to stay within API quotas
  • Type-Safe: Written in TypeScript with full type definitions

Quick Start

Option 1: npm (Recommended)

# Install globally
npm install -g @furkankoykiran/upwork-mcp

# Or use with npx (no installation needed)
npx @furkankoykiran/upwork-mcp

Option 2: With Claude Code (One-Line Setup)

claude mcp add upwork -- npx -y @furkankoykiran/upwork-mcp

Then set your credentials:

export UPWORK_CLIENT_ID="your_client_id"
export UPWORK_CLIENT_SECRET="your_client_secret"

And authenticate:

npx @furkankoykiran/upwork-mcp auth

Option 3: From Source

git clone https://github.com/furkankoykiran/upwork-mcp.git
cd upwork-mcp
npm install
npm run build

Authentication

1. Get API Credentials

  1. Go to Upwork Developer Portal
  2. Create a new API application
  3. Note your Client ID and Client Secret

2. Configure and Authenticate

# Set your credentials as environment variables
export UPWORK_CLIENT_ID="your_client_id"
export UPWORK_CLIENT_SECRET="your_client_secret"

# Run authentication
npx @furkankoykiran/upwork-mcp auth

This will:

  1. Open your browser with the Upwork OAuth authorization page
  2. Wait for you to log in and authorize the application
  3. Save your access token locally

3. Verify Authentication

npx @furkankoykiran/upwork-mcp check

Configuration

The server requires the following environment variables:

Variable Description Required
UPWORK_CLIENT_ID Your Upwork OAuth Client ID Yes
UPWORK_CLIENT_SECRET Your Upwork OAuth Client Secret Yes
UPWORK_API_URL Upwork GraphQL API URL (default: https://api.upwork.com/graphql) No

With Claude Code

Add to your MCP configuration (~/.config/claude-code/settings.json):

{
  "mcpServers": {
    "upwork": {
      "command": "npx",
      "args": ["-y", "@furkankoykiran/upwork-mcp"],
      "env": {
        "UPWORK_CLIENT_ID": "your_client_id",
        "UPWORK_CLIENT_SECRET": "your_client_secret"
      }
    }
  }
}

Available Tools

Profile Tools (4 tools)

Tool Description
get_profile Get your freelancer profile, skills, rate, and Job Success Score
get_profile_completeness Check profile completeness score and missing fields
get_skills List all skills on your profile
get_connects_balance Get your current connects balance

Job Search Tools (2 tools)

Tool Description
search_jobs Search job postings with filters (keyword, budget, category, etc.)
get_job_details Get full details of a specific job posting

Proposal Tools (6 tools)

Tool Description
list_proposals List your submitted proposals with status filter
get_proposal Get detailed proposal information including cover letter
submit_proposal Submit a new proposal (uses Connects)
update_proposal Update an existing proposal (pending only)
withdraw_proposal Withdraw a submitted proposal
get_proposal_stats Get proposal statistics and success rates

Saved Jobs Tools (3 tools)

Tool Description
list_saved_jobs List your saved/bookmarked jobs
save_job Save or unsave a job to bookmarks
get_job_recommendations Get personalized job recommendations

Contract Tools (2 tools)

Tool Description
list_contracts List your active and past contracts
get_contract_details Get detailed information about a specific contract

Analytics Tools (1 tool)

Tool Description
get_time_report Get time report for work history and hours logged

Total: 18 tools

Example Usage

Search for Jobs

Search for Upwork jobs with keyword "TypeScript" and hourly rate between $50-$100

Submit a Proposal (with Dry Run)

Submit a proposal for job ~0123456789012345678 with cover letter "I am experienced..." bid amount $75 hourly, dry_run=true

List Proposals

List my pending Upwork proposals

Save a Job

Save job ~0123456789012345678

Get Job Recommendations

Get personalized job recommendations

Safety Features

Dry Run Mode

All mutation tools (submit_proposal, update_proposal, withdraw_proposal) support a dry_run parameter that previews the action without executing it:

{
  "job_id": "~0123456789012345678",
  "cover_letter": "Your cover letter here...",
  "bid_amount": 50,
  "bid_type": "hourly",
  "dry_run": true
}

Validation

  • Cover letters must be 50-5000 characters
  • Bid amounts must be greater than 0
  • Proposal updates only work on pending proposals

Rate Limits

The server implements conservative rate limiting to avoid exceeding Upwork's API quotas:

  • 10 requests per second
  • 300 requests per minute
  • 40,000 requests per day

Rate limits are automatically enforced with token bucket algorithm.

Troubleshooting

Authentication Issues

# Clear existing token
npx @furkankoykiran/upwork-mcp logout

# Re-authenticate
npx @furkankoykiran/upwork-mcp auth

Token Expired

The server automatically attempts to refresh tokens when they expire. If this fails:

npx @furkankoykiran/upwork-mcp auth

Rate Limit Errors

If you hit rate limits, the server will automatically retry with exponential backoff. For persistent issues, wait a few minutes before retrying.

Development

# Clone the repository
git clone https://github.com/furkankoykiran/upwork-mcp.git
cd upwork-mcp

# Install dependencies
npm install

# Watch mode (rebuilds on changes)
npm run dev

# Run tests
npm test

# Lint code
npm run lint

# Format code
npm run format

See CONTRIBUTING.md for detailed contribution guidelines.

Project Structure

upwork-mcp/
├── src/
│   ├── index.ts              # MCP server entry point
│   ├── auth/                 # OAuth 2.0 authentication
│   ├── client/               # Upwork API client with rate limiting
│   ├── queries/              # GraphQL queries and mutations
│   ├── tools/                # MCP tool implementations
│   └── types/                # TypeScript type definitions
├── .github/
│   ├── workflows/            # CI/CD workflows (CI, CodeQL, Release, Stale)
│   ├── pull_request_template/
│   ├── ISSUE_TEMPLATE/       # Issue templates (Bug, Feature, Question)
│   ├── CONTRIBUTING.md       # Contribution guidelines
│   ├── CODE_OF_CONDUCT.md    # Community guidelines
│   ├── SECURITY.md           # Security policy
│   └── SUPPORT.md            # Support guidelines
├── dist/                     # Compiled output (gitignored)
├── node_modules/             # Dependencies (gitignored)
├── eslint.config.js          # ESLint configuration
├── tsconfig.json             # TypeScript configuration
├── package.json              # Project metadata
├── CHANGELOG.md              # Version history
├── LICENSE                   # MIT License
└── README.md                 # This file

License

MIT © Furkan Köykıran - see LICENSE file for details.

Links

Acknowledgments

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