Red/Blue Team UI
Learn how to use the Red/Blue Team interface for security analysis, vulnerability management, and follow-up questions.
Starting an Analysis
Select Red/Blue Team Mode
Click the mode selector below the chat input and choose "Red/Blue Team". This mode requires a Power tier subscription.
Paste Your Content
Enter the code, configuration, policy, contract, or other content you want analyzed. The system auto-detects the content type.
Review Results
The analysis appears with three tabs: Summary (overview and confidence), Vulnerabilities (detailed findings), and Hardened (fixed version).
Response Interface
The Red/Blue Team response has three main sections accessible via tabs:
Summary Tab
- • Confidence Score - Overall security assessment (0-100%)
- • Content Type - Detected type (code, config, policy, etc.)
- • Models Used - Red Team and Blue Team AI models
- • Statistics - Counts by severity level
- • Resolution Summary - Found, resolved, disputed counts
Vulnerabilities Tab
- • Filter Bar - Filter by All, Active, Resolved, or Disputed
- • Vulnerability Cards - Each finding with full details
- • Severity Badge - Color-coded severity level
- • Status Indicator - Active, Resolved, or Disputed
- • Expand/Collapse - Click cards to see full details
Hardened Tab
- • Fixed Content - Security-hardened version ready to use
- • Copy Button - One-click copy to clipboard
- • Markdown Rendering - Proper code formatting and syntax
Understanding Vulnerability Cards
Each vulnerability is displayed as an expandable card with comprehensive information:
ID: VULN-001
Category: Injection
Location: Line 15, get_user() function
Description: User input directly interpolated into SQL query...
Impact: Attackers can read, modify, or delete database records...
Suggested Fix: Use parameterized queries with prepared statements...
Asking Follow-up Questions
After an analysis completes, you can ask detailed questions about the findings.
Follow-up Mode
- 1. Click "Ask About This" button on the analysis response
- 2. A green indicator appears below the input showing you're in follow-up mode
- 3. Type your question - the AI has full context of the analysis
- 4. Click "Exit Q&A" to return to normal chat
Example Follow-up Questions
- • "Why is VULN-002 marked as a false positive?"
- • "Explain the XSS vulnerability in more detail"
- • "How would an attacker exploit this SQL injection?"
- • "What's the best way to implement the suggested fix in Node.js?"
- • "Are there any other vulnerabilities I should look for?"
- • "Why is the confidence score 75%?"
Vulnerability Statuses
Tips for Best Results
Provide Complete Context
Include surrounding code or full configuration files when possible. The more context, the more accurate the analysis.
Review Disputed Findings
When Blue Team disputes a finding, read the reasoning carefully. Some disputes may still warrant attention.
Use Follow-ups for Clarification
If a vulnerability or fix isn't clear, use follow-up questions. Reference specific vulnerability IDs for precise answers.
Verify Hardened Output
Always review the hardened output before using it. AI-generated fixes should be tested in your environment.