Cv Analysis

For DOCX files:

• Use textutil (macOS built-in): textutil -convert txt -stdout "FILE_PATH"
• Or use python3 with python-docx if textutil output is insufficient

CV Analysis Process

After extracting the CV content: 2. Extract and categorize:

• Education: degrees, institutions, dates, grades (note: do NOT judge grades, just record them)
• Technical Skills: programming languages, frameworks, tools
• Project Experience: name, description, technologies used, achievements — flag which projects are most relevant to research applications
• Work Experience: job titles, companies, responsibilities
• Internships: similar to work experience
• Awards & Honors: competitions, scholarships, hackathons
• Teaching Experience: courses taught, role (TA/instructor)
• Publications: papers, venues (if any)

3. Evaluate for research application relevance:
   • Strengths: experiences, skills, and achievements that strengthen a research degree application
   • Areas for Improvement: gaps that could be addressed (no publications, low grades, etc.) — note these objectively without judgment
   • Packaging Opportunities: experiences that could be framed more strongly for applications

Output

Write the structured profile to ${CLAUDE_PLUGIN_ROOT}/.local.md:

• Set cv_profile_analyzed: true in frontmatter
• Set cv_file_path in frontmatter if not already set
• Populate the ## CV Profile markdown section with all extracted information
• Use clear headers and bullet points

After writing, present a summary to the user highlighting:

• Top 3 strengths for research applications
• Key experiences that connect to potential research directions
• Any notable packaging opportunities

Important Notes

• Never judge or comment negatively on grades — they are factual data
• Focus on identifying connections between experience and research potential
• The output must be structured enough for other skills (cold-email, professor-match, etc.) to reference programmatically
• Support both English and Chinese CVs