Upwork MCP Server
mcp-name: io.github.wmarceau/upwork
A Model Context Protocol (MCP) server for Upwork job discovery, client analysis, and proposal assistance. Helps freelancers find relevant opportunities and craft winning proposals.
Features
- Job Search - Search Upwork marketplace with keyword, budget, and client quality filters
- Job Details - Get full job descriptions, requirements, and screening questions
- Client Analysis - Analyze client reliability, hire rate, and spending history
- Proposal Drafting - Generate customized proposal drafts based on job requirements
- Profile Management - View your profile, stats, and portfolio
- Proposal Tracking - Track submitted proposals and their status
- Contract History - View active and past contracts with earnings
Installation
pip install upwork-mcp
Setup
1. Get Upwork API Credentials
- Go to Upwork Developer Portal
- Create a new API application
- Note your
client_idandclient_secret
2. Configure Environment Variables
export UPWORK_CLIENT_ID="your_client_id"
export UPWORK_CLIENT_SECRET="your_client_secret"
export UPWORK_REDIRECT_URI="https://localhost" # or your callback URL
3. Add to Claude Desktop
Add to your Claude Desktop configuration (~/.claude/claude_desktop_config.json):
{
"mcpServers": {
"upwork": {
"command": "upwork-mcp",
"env": {
"UPWORK_CLIENT_ID": "your_client_id",
"UPWORK_CLIENT_SECRET": "your_client_secret"
}
}
}
}
4. Authenticate
On first use, run the upwork_auth_status tool to get the authorization URL. Visit the URL, authorize the app, and use upwork_complete_auth with the callback URL.
Available Tools
Job Discovery
upwork_search_jobs
Search for jobs matching your skills and preferences.
Input:
- query: Search keywords (required)
- min_budget: Minimum budget filter
- max_budget: Maximum budget filter
- payment_verified: Only verified clients (default: true)
- limit: Max results (default: 20)
Output: Job listings with title, budget, client info, skills
upwork_get_job_details
Get full details about a specific job.
Input:
- job_id: Upwork job ID (required)
Output: Full description, questions, attachments, client details
Client Intelligence
upwork_analyze_client
Analyze a client's reliability and hiring patterns.
Input:
- job_id: Job ID to analyze client from (required)
Output: Hire rate, spend history, red flags, recommendation
Proposal Assistance
upwork_draft_proposal
Generate a customized proposal draft.
Input:
- job_id: Job ID to write proposal for (required)
- highlight_skills: Skills to emphasize
- relevant_experience: Experience to mention
- proposed_rate: Your rate
Output: Cover letter template, question answers, rate
Note: Proposals must be submitted manually. Automated submission is not allowed by Upwork.
Profile & Tracking
upwork_get_my_profile
View your freelancer profile and stats.
upwork_get_my_proposals
Track your submitted proposals.
Input:
- status: PENDING, INTERVIEWING, ARCHIVED, or ALL
- limit: Max results
upwork_get_my_contracts
View your contracts and earnings.
Input:
- status: ACTIVE, ENDED, or ALL
- limit: Max results
Authentication
upwork_auth_status
Check authentication status and get auth URL if needed.
upwork_complete_auth
Complete OAuth flow with callback URL.
Example Usage
Finding Jobs
User: Search for Amazon FBA analytics jobs with budget over $500
Claude: [Uses upwork_search_jobs with query="amazon fba analytics" min_budget=500]
Evaluating Opportunities
User: Analyze the client for job 123456
Claude: [Uses upwork_analyze_client]
Result shows:
- Client has posted 45 jobs, hired 38 (84% hire rate)
- Total spend: $125,000
- Payment verified
- Rating: LOW RISK ✅
Drafting Proposals
User: Draft a proposal for job 123456, emphasizing my Python and SP-API experience
Claude: [Uses upwork_draft_proposal with highlight_skills and experience]
Rate Limits
- Upwork API: 300 requests/minute per IP
- GraphQL queries are batched where possible
Privacy & Security
- OAuth tokens are stored locally at
~/.upwork/token.json - No data is sent to third parties
- API requests go directly to Upwork
License
MIT License - see LICENSE
Support
- Issues: GitHub Issues
- Upwork API Docs: developer.upwork.com
Metadata
Release files for upwork-mcp 1.0.1
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| upwork_mcp-1.0.1.tar.gz | 12.4 kB | Without provenance | Details for upwork_mcp-1.0.1.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| upwork_mcp-1.0.1-py3-none-any.whl | Python 3 | none | any | Details for upwork_mcp-1.0.1-py3-none-any.whl |
Total release size: 25.4 kB
Release files / upwork_mcp-1.0.1.tar.gz
| Download URL | upwork_mcp-1.0.1.tar.gz |
|---|---|
| Size | 12.4 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
651a0bba23a0eabbc38cffd723e21d01ddab5fa4cc045da26c592c97212c6231
|
|
BLAKE2b-256 checksum How to use checksums |
796a0a7e9144e1f954b117e46dde81f40bbfe7e81f8bd47cb34191baee738e4c
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.13.5
|
Release files / upwork_mcp-1.0.1-py3-none-any.whl
| Download URL | upwork_mcp-1.0.1-py3-none-any.whl |
|---|---|
| Size | 12.9 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
a25cd229e88cb00150701d2c11ff3e45c38173afc53eb0fe0d25930a3e0cb703
|
|
BLAKE2b-256 checksum How to use checksums |
1c13885e4c4ea704bb3ee64c0e0c191e2262da4894bec12beedf0ce3a90c3d23
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.13.5
|
Security for upwork-mcp 1.0.1
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.