[P3-Feature] Build MCP server to expose ResumeFlow as AI assistant tools

[Ztrimus]

Overview

Expose the ResumeFlow pipeline as an MCP (Model Context Protocol) server. This makes it usable from Claude Desktop, VS Code Copilot, Cursor, or any MCP-compatible AI assistant.

MCP Tools to expose

parse_job_description(url: str) → JobDetails
parse_resume(file_path: str) → ResumeData
tailor_resume(job_details: JobDetails, resume: ResumeData) → TailoredResume
generate_pdf(resume: TailoredResume) → file_path
generate_cover_letter(job_details, resume) → str
score_match(resume: str, jd: str) → MatchScore

Impact

• Users can say "tailor my resume for this job posting" in Claude Desktop
• Integrates with any AI coding assistant that supports MCP
• Turns ResumeFlow into a composable building block

References

• MCP spec: https://modelcontextprotocol.io
• Python SDK: mcp package