koachgg/mcp-resume-server
If you are the rightful owner of mcp-resume-server and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to dayong@mcphub.com.
This is an MCP (Model Context Protocol) server implementation for submitting resumes to Puch AI.
MCP Server for Resume Submission
This is an MCP (Model Context Protocol) server implementation for submitting your resume to Puch AI.
Setup Instructions
1. Prerequisites
- Python 3.8 or higher
- Your resume file in one of these formats: PDF, DOCX, DOC, TXT, or MD
2. Installation
- Create and activate a virtual environment:
python -m venv venv
.\venv\Scripts\Activate.ps1
- Install dependencies:
pip install -r requirements.txt
3. Configuration
-
Update your application key: Replace
<generated_token>inmcp_server.pywith your actual application key from Puch. -
Update your phone number: Replace
9189XXXXXXXXinmcp_server.pywith your phone number in the format{country_code}{number}(without the + symbol).- Example:
919876543210for an Indian number
- Example:
-
Add your resume: Place your resume file in the same directory as
mcp_server.py. The server will automatically detect files with these names:resume.pdf,resume.docx,resume.doc,resume.txt,resume.mdcv.pdf,cv.docx,cv.doc,cv.txt,cv.md- Any file containing "resume" or "cv" in the filename
4. Running the Server
python mcp_server.py
The server will start on http://0.0.0.0:8085
5. Making the Server Publicly Accessible
For Puch to connect to your server, it needs to be publicly accessible. You can use:
-
ngrok (recommended for testing):
ngrok http 8085This will give you a public URL like
https://abc123.ngrok.io -
Cloudflare Tunnel
-
Your own VPS/cloud server
6. Connecting with Puch
Use the command:
/mcp connect <YOUR_PUBLIC_SERVER_URL>/mcp <YOUR_AUTH_TOKEN>
Example:
/mcp connect https://abc123.ngrok.io/mcp your_application_key_here
Features
The server implements three tools as required:
- resume(): Returns your resume in markdown format
- validate(): Returns your phone number for validation
- fetch(): Fetches web content (utility tool)
Supported Resume Formats
- PDF: Extracts text using PyPDF2
- DOCX: Converts to markdown using mammoth (preferred) or docx2txt (fallback)
- DOC: Converts using docx2txt
- TXT: Reads as plain text
- MD: Reads as markdown
Troubleshooting
- Resume not found: Make sure your resume file is in the same directory and has a recognizable name
- Port in use: Change the port number in
mcp_server.pyif 8085 is already occupied - Import errors: Make sure all dependencies are installed with
pip install -r requirements.txt
Security Note
This server uses a simple bearer token authentication. In production, consider implementing more robust authentication mechanisms.