CV Resume Builder MCP

Instructions

Implementation Reference

• src/cv_resume_builder_mcp/server.py:886-975 (handler)

  The handler function that executes the generate_enhanced_cv tool. It parses the existing CV (PDF or LaTeX), gathers recent data from git logs, wins.md, Jira tickets, Credly badges, LinkedIn config, and compiles a comprehensive report with guidelines for enhancing the CV.
  async def generate_enhanced_cv(existing_cv_path: Optional[str], since: str) -> list[TextContent]: """Generate enhanced CV combining all data sources.""" if not existing_cv_path: return [TextContent(type="text", text="Error: existingCvPath parameter is required")] # Resolve path if existing_cv_path.startswith('/'): resolved_path = Path(existing_cv_path) else: resolved_path = Path(REPO_PATH) / existing_cv_path if not resolved_path.exists(): return [TextContent(type="text", text=f"CV file not found: {resolved_path}")] try: # Step 1: Parse existing CV if resolved_path.suffix.lower() == '.pdf': reader = pypdf.PdfReader(str(resolved_path)) existing_content = "" for page in reader.pages: existing_content += page.extract_text() + "\n" else: existing_content = resolved_path.read_text() # Step 2: Gather all data sources data_sections = [] # Git commits git_result = await get_git_log(since) data_sections.append(f"\n## Git Commits ({since}):\n{git_result[0].text}") # Wins wins_result = await read_wins() data_sections.append(f"\n## Achievements (wins.md):\n{wins_result[0].text}") # Jira tickets if all([JIRA_URL, JIRA_EMAIL, JIRA_API_TOKEN]): jira_result = await get_jira_tickets(since, "Done") data_sections.append(f"\n{jira_result[0].text}") else: data_sections.append("\n## Jira Tickets: Not configured") # Credly badges if CREDLY_USER_ID: credly_result = await get_credly_badges(50) data_sections.append(f"\n{credly_result[0].text}") else: data_sections.append("\n## Credly Badges: Not configured") # LinkedIn if LINKEDIN_PROFILE_URL: data_sections.append(f"\n## LinkedIn Profile: {LINKEDIN_PROFILE_URL}\n(Manual review recommended)") # Step 3: Compile report truncated_cv = existing_content[:2000] if len(existing_content) > 2000: truncated_cv += "\n... (truncated, full content available)" report = f"""# Enhanced CV Generation Report ## Existing CV Content: {truncated_cv} ## Recent Work & Achievements ({since}): {''.join(data_sections)} --- ## CV Guidelines: - Maximum {MAX_BULLETS_PER_EXPERIENCE} bullet points per experience - Focus on quantifiable impact and results - Use action verbs (Engineered, Designed, Implemented, Led, etc.) - Include metrics (percentages, time savings, scale) ## Next Steps: 1. Review the existing CV content above 2. Analyze the recent work data (git commits, Jira tickets, badges) 3. Identify the top {MAX_BULLETS_PER_EXPERIENCE} most impactful achievements 4. Generate enhanced CV bullet points in LaTeX format 5. Ensure each bullet demonstrates clear business value ## Suggested Approach: Ask the AI to: "Based on the above data, generate {MAX_BULLETS_PER_EXPERIENCE} enhanced CV bullet points in LaTeX format that highlight my most significant recent achievements with quantifiable results." """ return [TextContent(type="text", text=report)] except Exception as e: return [TextContent(type="text", text=f"Enhanced CV generation error: {str(e)}")]
• src/cv_resume_builder_mcp/server.py:282-300 (registration)

  Registers the generate_enhanced_cv tool in the MCP server's list_tools() function, including name, description, and input schema.
  Tool( name="generate_enhanced_cv", description="Generate an enhanced CV by combining existing CV content with data from all sources", inputSchema={ "type": "object", "properties": { "existingCvPath": { "type": "string", "description": "Path to existing CV (PDF or LaTeX)" }, "since": { "type": "string", "description": "Time range for fetching recent work", "default": "6 months ago" } }, "required": ["existingCvPath"] } )
• src/cv_resume_builder_mcp/server.py:285-299 (schema)

  Input schema definition for the generate_enhanced_cv tool, specifying required existingCvPath and optional since parameter.
  inputSchema={ "type": "object", "properties": { "existingCvPath": { "type": "string", "description": "Path to existing CV (PDF or LaTeX)" }, "since": { "type": "string", "description": "Time range for fetching recent work", "default": "6 months ago" } }, "required": ["existingCvPath"] }