AI CV Screening & Candidate Scoring

CV Screening Workflow Setup Guide

What This Agent Does

The CV Screening Workflow automates the entire candidate evaluation process by collecting job applications, analyzing CVs with AI intelligence, and maintaining a ranked candidate database. When candidates submit applications through your form, their CVs are automatically uploaded, analyzed using advanced AI scoring, and organized in a Google Sheet sorted by qualification score—transforming hours of manual screening into a streamlined, objective process.

Key benefits include:

• 90% time savings on initial CV screening and candidate ranking
• Objective evaluation using consistent AI-powered criteria across all candidates
• Real-time candidate database automatically updated and sorted by qualification score
• Audit trail of all evaluations with detailed scoring rationale
• Scalability to handle hundreds of applications without additional effort

Target use cases: Recruiting teams managing high-volume applications, HR departments standardizing candidate evaluation, talent acquisition teams needing rapid initial screening, and organizations seeking to reduce hiring bias through objective AI analysis.

Who Is It For

This workflow is designed for:

• Recruiting professionals who spend significant time on initial CV screening
• HR teams managing multiple open positions simultaneously
• Hiring managers who need objective candidate rankings before interviews
• Startups and growing companies lacking dedicated recruiting resources
• Enterprise organizations processing hundreds of applications monthly

No technical expertise required—the workflow handles all automation once configured.

Required Integrations

Google Drive

Why it's needed: Google Drive stores uploaded CVs securely and provides the document source for AI analysis. This integration ensures candidates' application documents are preserved and accessible throughout the evaluation process.

Setup steps:

1. Log into your Google Cloud Console at console.cloud.google.com
2. Create a new project or select an existing one
3. Enable the Google Drive API:
   • Navigate to "APIs & Services" → "Library"
   • Search for "Google Drive API"
   • Click "Enable"
4. Create OAuth 2.0 credentials:
   • Go to "APIs & Services" → "Credentials"
   • Click "Create Credentials" → "OAuth client ID"
   • Select "Web application"
   • Add authorized redirect URI: https://app.taskagi.com/oauth/callback
   • Copy your Client ID and Client Secret
5. In TaskAGI, navigate to Integrations → Google Drive
6. Paste your Client ID and Client Secret
7. Click Authorize and grant TaskAGI access to your Google Drive
8. Create a dedicated folder in Google Drive named CV_Submissions for organized storage

Configuration in TaskAGI:

• Set the upload destination folder to your CV_Submissions folder
• Ensure the connected Google account has write permissions to this folder

Google Gemini (AI Analysis)

Why it's needed: Google Gemini's advanced language model analyzes CVs against your job requirements, scoring candidates objectively on skills, experience, and qualifications. This eliminates manual reading and provides consistent, bias-reduced evaluation.

Setup steps:

1. Visit ai.google.dev and sign in with your Google account
2. Navigate to API Keys section
3. Click Create API Key
4. Copy the generated API key
5. In TaskAGI, go to Integrations → Google Gemini
6. Paste your API key in the configuration field
7. Verify the model is set to gemini-2.5-flash (recommended for speed and accuracy)
8. Test the connection by clicking Test Integration

How to obtain credentials:

• Google Gemini API keys are free to generate and include generous free-tier usage
• No billing setup required for initial testing
• Visit ai.google.dev/pricing to understand usage limits

Configuration in TaskAGI:

• Model: gemini-2.5-flash
• Temperature: 0.3 (for consistent, objective scoring)
• Max tokens: 1000 (sufficient for detailed analysis)

Google Sheets

Why it's needed: Google Sheets serves as your candidate database, storing evaluation results, scores, and rankings in an accessible, shareable format. This creates a single source of truth for your hiring team.

Setup steps:

1. Create a new Google Sheet at sheets.google.com
2. Name it Candidate_Evaluations (or your preferred name)
3. Set up column headers in the first row:
   • Column A: Candidate_Name
   • Column B: Email
   • Column C: Position_Applied
   • Column D: CV_Link
   • Column E: Overall_Score
   • Column F: Skills_Match
   • Column G: Experience_Level
   • Column H: Evaluation_Summary
4. Share the sheet with your hiring team (set permissions as needed)
5. Copy the sheet URL from your browser address bar
6. In TaskAGI, go to Integrations → Google Sheets
7. Authorize access using the same Google account as your Drive setup
8. Paste your sheet URL in the configuration field

Configuration in TaskAGI:

• Sheet URL: https://docs.google.com/spreadsheets/d/YOUR_SHEET_ID/edit
• Worksheet name: Sheet1 (or your active sheet name)
• Append mode: Enabled (to add new candidates without overwriting)

Configuration Steps

Step 1: Configure the Job Application Form Trigger

The workflow begins with your application form. This node collects candidate information and CV uploads.

Configuration:

• Form type: Select "Google Form" or "Custom Form"
• Required fields:
  • Candidate Full Name
  • Email Address
  • Position Applied For
  • CV Upload (file type: PDF or DOCX)
• Optional fields:
  • Phone Number
  • LinkedIn Profile
  • Cover Letter

Example form structure:

1. Full Name [text field]
2. Email [email field]
3. Position [dropdown: Senior Developer, Junior Developer, etc.]
4. Upload CV [file upload - PDF/DOCX only]

Step 2: Configure CV Upload to Google Drive

This node automatically saves submitted CVs to your Drive folder.

Parameters:

• Destination folder: CV_Submissions
• File naming: {Candidate_Name}_{Date}_{Position}
• Example output: John_Smith_2024-01-15_Senior_Developer.pdf

Data mapping:

• Source: Form submission file
• Destination: Google Drive folder path
• Output: CV file URL (used in next steps)

Step 3: Configure Prepare Candidate Data

This processing node structures form data for AI analysis.

Function logic:

Input: Form submission data
Output: Structured object containing:
  - candidate_name
  - email
  - position
  - cv_url
  - submission_date

No manual configuration needed—this node automatically formats data from the form trigger.

Step 4: Configure AI CV Analysis & Scoring

This is the core intelligence node. It analyzes CVs against job requirements.

Critical configuration:

The analysis prompt is pre-configured but should be customized for your roles:

You are an expert HR analyst evaluating a job candidate. 
Analyze the provided CV against these criteria:
- Technical skills match (0-25 points)
- Relevant experience (0-25 points)
- Education and certifications (0-20 points)
- Career progression (0-15 points)
- Cultural fit indicators (0-15 points)

Provide:
1. Overall score (0-100)
2. Breakdown by category
3. Key strengths
4. Potential gaps
5. Recommendation (Strong Yes/Yes/Maybe/No)

Customization tips:

• Modify scoring weights based on your role priorities
• Add specific technical skills your position requires
• Include industry-specific evaluation criteria
• Adjust recommendation thresholds for your hiring standards

Parameters:

• Model: gemini-2.5-flash
• Temperature: 0.3 (consistent scoring)
• Document source: CV URL from Drive upload

Step 5: Configure Parse AI Results

This node extracts structured data from AI analysis.

Output structure:

{
  "overall_score": 85,
  "skills_match": 90,
  "experience_level": 80,
  "recommendation": "Strong Yes",
  "summary": "Excellent technical fit with 8 years relevant experience..."
}

No configuration needed—automatically parses Gemini output.

Step 6: Configure Prepare Sheet Data

This node formats all information for Google Sheets storage.

Data mapping:

• Candidate Name → Column A
• Email → Column B
• Position Applied → Column C
• CV Link → Column D
• Overall Score → Column E
• Skills Match → Column F
• Experience Level → Column G
• Evaluation Summary → Column H

Step 7: Configure Save to Google Sheets

Appends candidate evaluation to your sheet.

Critical setup:

• Sheet URL: Paste your Google Sheet URL here
• Worksheet name: Sheet1 (or your active sheet)
• Append mode: Enabled
• Header row: Row 1 (contains column names)

Step 8: Configure Get All Candidates

Retrieves all existing candidate records for sorting.

Parameters:

• Sheet URL: Same as Step 7
• Range: A:H (all columns with data)
• Include headers: Yes

Step 9: Configure Sort by Score

Processes candidate list to rank by evaluation score.

Sort logic:

• Primary sort: Overall_Score (descending—highest first)
• Secondary sort: Submission_Date (ascending—earliest first)

Output: Sorted candidate array ready for sheet update

Step 10: Configure Clear Sheet & Write Sorted Data

These nodes clear old data and write the newly sorted candidate list.

Clear Sheet parameters:

• Range: A2:H1000 (preserves headers, clears data)
• Keep headers: Yes

Write Sorted Data (Loop):

• Loop through: Sorted candidates array
• Action per iteration: Append row to Google Sheets
• Data source: Current candidate object from loop

Testing Your Agent

Test Execution Steps

1. Submit a test application:

• Fill out your job application form completely
• Upload a sample CV (use a real CV or create a test document)
• Submit the form
• Verify: Form submission completes without errors

2. Monitor CV upload:

• Check your CV_Submissions folder in Google Drive
• Expected result: New CV file appears within 30 seconds
• Verify: File naming follows {Name}_{Date}_{Position} pattern

3. Check AI analysis:

• Wait 60 seconds for Gemini analysis to complete
• Review TaskAGI execution logs for the AI node
• Expected result: Detailed scoring breakdown appears in logs
• Verify: Score is between 0-100 with category breakdowns

4. Validate Google Sheets update:

• Open your Candidate_Evaluations sheet
• Expected result: New row appears with all candidate data
• Verify: Score, summary, and all fields are populated correctly

5. Test sorting functionality:

• Submit 2-3 additional test applications
• Wait for all to process
• Expected result: Sheet automatically re-sorts with highest scores at top
• Verify: Candidates appear in descending score order

Success Indicators

✅ Workflow is working correctly when:

• All form submissions trigger the workflow automatically
• CVs appear in Google Drive within 30 seconds
• AI analysis completes within 60 seconds per CV
• Google Sheets updates with new candidates automatically
• Candidate list remains sorted by score after each new submission
• No error messages in TaskAGI execution logs
• All data fields populate correctly in the sheet

Troubleshooting Common Issues

You're now ready to automate your CV screening process! Start with a test submission, monitor the execution, and adjust the AI prompt based on your first results. Your hiring team will immediately benefit from objective, rapid candidate evaluation.