Cv Manager

CV Manager

Overview

Manage candidate CVs and interview records in a structured Google Drive/Docs system. Store CVs in a dedicated Drive folder and maintain a master tracking document with candidate profiles and interview summaries.

Core Capabilities

This skill provides three main workflows:

1. Add a CV - Upload and process new candidate CVs
2. Add an Interview - Record and structure interview notes
3. Summarize a Candidate - Generate comprehensive candidate summaries

Configuration

Reference: references/configuration.md for Google Drive folder IDs and document IDs.

Key Resources:

• CV Storage Folder: 1n5sMW4L_NHpnIFVogRD2I3jPvKk2WQUw
• Interview Files Folder: 1fRRtK3TPxbTDSPw_KufOrYLHMNuelLp0 (audio recordings, attachments)
• Tracking Document: 1M83HrxnWIOYUGk_jtaaLS_eo5_-sIlHmi_1ZAwCdRWI

Document Structure

Reference: references/document-structure.md for detailed structure template.

Master Document Structure:

• Heading 1 (Title): CVs
• Heading 2: Table of content

Per-Candidate Hierarchy:

• Heading 2: Candidate Name
• Heading 3: Presentation (CV info, job title, 100-200 word description)
• Heading 3: Interview
• Heading 4: Interview Date
• Heading 5: Attendees
• Heading 5: Summary (interview content)

Note: Each candidate is a Heading 2 section within the master tracking document.

Workflow 1: Add a CV

Trigger: User provides a local CV file (PDF, DOCX, etc.) and candidate name.

Steps:

1. Receive the CV file

   • User provides local file path to CV
   • Confirm candidate name

2. Extract CV information

   • Use scripts/run.sh extract_cv <file_path> to extract text from PDF
   • Use pdf-extractor skill for complex PDFs if needed
   • Analyze extracted text to identify:
     • Job title/role
     • Key skills and expertise
     • Experience highlights
     • Notable achievements

3. Generate presentation content

   • Draft Job Title summary
   • Write 100-200 word description with main skills
   • Present draft to user for review
   • If user requests, run open <file_path> to view original CV

4. Upload CV to Google Drive

   • Use google-drive-manager skill to upload file to CV Storage Folder
   • Folder ID: 1n5sMW4L_NHpnIFVogRD2I3jPvKk2WQUw
   • File name format: <Candidate Name> - CV.pdf
   • Capture the returned Google Drive link

5. Get formal validation

   • Present complete entry to user:
     • Candidate name (Heading 2)
     • Presentation section with CV link, job title, description
   • Request approval before insertion

6. Insert into tracking document

   • Use google-docs-manager skill to:
     • Create new Heading 2 with candidate name
     • Add Presentation section (Heading 3)
     • Insert CV Link, Job Title, and description
     • Create empty Interview section (Heading 3)
   • Document ID: 1M83HrxnWIOYUGk_jtaaLS_eo5_-sIlHmi_1ZAwCdRWI
   • Insert after "Table of content" section

7. Cleanup

   • Remove local CV file using rm <file_path>
   • Confirm completion to user

Workflow 2: Add an Interview

Trigger: User provides interview notes and/or audio recordings for existing candidate.

Steps:

1. Identify candidate

   • User provides candidate name
   • Use google-docs-manager skill to verify candidate exists in tracking document
   • Locate their Interview section (Heading 3)

2. Upload interview files to Drive (if provided)

   • If audio file or other attachments provided:
     • Use google-drive-manager skill to upload file to Interview Files Folder
     • Folder ID: 1fRRtK3TPxbTDSPw_KufOrYLHMNuelLp0
     • Files are backed up for future reference

3. Process interview content

   • If audio file provided:
     • Use speech-to-text skill to transcribe audio
     • Save transcription as interview notes
   • If text notes provided:
     • Use notes directly

4. Extract interview information

   • Identify interview date
   • Extract or confirm attendee list
   • Present attendee list to user for confirmation
   • Draft interview summary including:
     • First paragraph: Global feeling, strengths, weaknesses
     • Main content: Key discussion points, technical assessments, notable responses

5. Get formal validation

   • Present complete interview entry:
     • Interview date (Heading 4)
     • Attendees section (Heading 5) with plain text list
     • Summary section (Heading 5)
     • Overall assessment paragraph
     • Main interview discussion summary
   • Request user to provide final thoughts
   • Incorporate user feedback and re-present if needed

6. Insert into document

   • Use google-docs-manager skill to:
     • Navigate to candidate's Interview section (Heading 3)
     • Add new Heading 4 with interview date
     • Add Heading 5 "Attendees" with plain text list (not bullets)
     • Add Heading 5 "Summary"
     • Insert overall assessment paragraph
     • Insert main interview discussion summary
   • Document ID: 1M83HrxnWIOYUGk_jtaaLS_eo5_-sIlHmi_1ZAwCdRWI

7. Cleanup

   • Automatically remove local interview files using rm <file_path>
   • Files are already backed up in Google Drive (Folder: 1fRRtK3TPxbTDSPw_KufOrYLHMNuelLp0)
   • Confirm completion to user

Workflow 3: Summarize a Candidate

Trigger: User requests summary or information about a specific candidate.

Steps:

1. Locate candidate information

   • User provides candidate name
   • Use google-docs-manager skill to read tracking document
   • Document ID: 1M83HrxnWIOYUGk_jtaaLS_eo5_-sIlHmi_1ZAwCdRWI
   • Find candidate's Heading 2 section

2. Extract all information

   • Read candidate's Heading 2 section
   • Read Presentation section (Heading 3):
     • CV link
     • Job title
     • Description
   • Read all Interview sections (Heading 3):
     • Interview dates (Heading 4)
     • Attendees (Heading 5)
     • Summary content (Heading 5)
     • Overall assessments
     • Discussion points

3. Generate comprehensive summary

   • Candidate overview (from Presentation)
   • Interview history:
     • Number of interviews conducted
     • Dates and attendees
     • Key feedback from each interview
     • Overall strengths noted across interviews
     • Areas of concern noted across interviews
   • Recommendation or next steps (if applicable from interviews)

4. Present summary to user

   • Formatted, clear summary with all relevant details
   • Include CV link for reference

Resources

scripts/

run.sh - Generic script runner using uv for isolated Python environments

• Usage: ./run.sh <script_name> [args...]
• Automatically loads environment variables from .env files
• Creates isolated Python environments for dependencies

extract_cv.py - Extract text from CV files

• Usage: scripts/run.sh extract_cv <file_path>
• Supports: PDF files
• Returns: Extracted text to stdout

references/

configuration.md - Google Drive and Docs IDs, authentication setup document-structure.md - Detailed template and example of document structure

Integration with Other Skills

• pdf-extractor: Use for complex PDF extraction when extract_cv.py fails
• google-docs-manager: Primary tool for reading/writing tracking document
• google-drive-manager: Alternative for Drive operations if needed
• speech-to-text: Transcribe interview audio recordings
• email-extractor: Extract CV from email attachments if sent via email

Important Notes

• Always get formal validation from user before inserting content into tracking document
• Respect the document structure hierarchy (Heading 2-5 for candidates) strictly
• One candidate can have multiple interviews but only one Presentation section
• Attendees are listed as plain text (not bullets) under the Attendees heading
• Summary section contains two paragraphs: overall assessment first, then detailed discussion
• Always insert new candidates after "Table of content" section
• Clean up local files after upload to Drive
• Use open <file> command when user wants to review the local CV before processing
• Each candidate is a Heading 2 section within the master tracking document