cv-resume-builder

This MCP server powers an AI-assisted CV and resume builder that compiles data from your Git commits, Jira tickets, Credly certifications, and existing CVs to generate ATS-friendly LaTeX resumes. The tools exposed by the server include capabilities to fetch Git history (get_git_log, get_git_log_by_repo, get_git_log_all_repos), analyze commits for impact (get_commit_details, analyze_commits_impact), read and parse existing CVs (read_cv, parse_cv_pdf), aggregate achievements (read_wins, get_credly_badges, get_linkedin_profile), and produce a consolidated, enhanced CV (generate_enhanced_cv) along with formatting guidelines (get_cv_guidelines). To use it, configure your MCP client (Claude Desktop, Kiro, or another MCP-enabled assistant) with the cv-resume-builder endpoint and run the uvx-based command shown in the configuration. You can then issue natural language prompts like “Get my git commits from the last 6 months and suggest CV updates” or “Generate an enhanced CV using my existing resume.pdf and recent work from the last 3 months” to drive the workflow. Tools will respond with structured data or trigger LaTeX CV generation, returning a ready-to-export resume in ATS-friendly format. If you want to test locally or develop features, you can reference the TESTING.md for local development flows and the available commands listed in the README’s tool table.

Prerequisites:

• Python 3.10+ installed
• Access to uvx (for running MCP servers) or Python development setup
• Optional: Git, Jira, Credly accounts for integrations

Install steps (recommended for production with uvx):

1. Install uvx if not already available and ensure Python is accessible in PATH.
2. Install the CV Resume Builder MCP package: pip3 install cv-resume-builder-mcp
3. If developing locally, clone the repository and install in editable mode: git clone https://github.com/eyaab/cv-resume-builder-mcp.git cd cv-resume-builder-mcp pip install -e ".[dev]"
4. Configure the MCP client to use uvx with the server name and environment variables as described in the README. For example, place a configuration similar to: { "mcpServers": { "cv-resume-builder": { "command": "uvx", "args": ["cv-resume-builder-mcp"], "env": { "AUTHOR_NAME": "your-git-username", "REPOS": "default:/absolute/path/to/your-repo" } } } }
5. Restart the MCP client or IDE integration to load the new server configuration.

Usage tip: The instructions favor uvx for automatic version retrieval from PyPI, but local development can opt into using python -m for local runs as noted in TESTING.md.

Tips and notes:

• The recommended runtime is uvx to fetch the latest stable version from PyPI. For local development or unreleased features, refer to TESTING.md.
• Env vars: AUTHOR_NAME should be your git username; REPOS should specify absolute paths to repositories, with multiple repos separated as CompanyA:/path1,CompanyB:/path2,Personal:/path3.
• Optional integrations include Jira (configure JIRA_URL, JIRA_EMAIL, JIRA_API_TOKEN, JIRA_USER), Credly (CREDLY_USER_ID), and LinkedIn (LINKEDIN_PROFILE_URL). Note that LinkedIn integration is not implemented yet; the server can still read the profile URL and guide you to copy achievements manually when needed.
• CV formatting can be customized via MAX_BULLETS_PER_EXPERIENCE to control bullet limits per experience entry.
• If you encounter MCP server visibility issues, verify that the URI and environment paths are absolute and that the MCP client has permissions to access the configured repositories.