Reports

Strix generates comprehensive reports in multiple formats for both human review and automated processing.

Output Directory Structure

Each scan creates a timestamped directory in strix_runs/:

strix_runs/
└── example_20260301_123456/
    ├── vulnerabilities.json          # Machine-readable vulnerability data
    ├── vulnerabilities.md             # Human-readable vulnerability reports  
    ├── scan_metadata.json             # Scan configuration and metadata
    ├── agent_traces.json              # Agent execution traces
    └── tool_executions.json           # Tool call history

Run Name Format

Run names follow the pattern: <target>_<date>_<time>

example_com_20260301_123456
github_user_repo_20260301_140530
multi_target_20260301_153245

Vulnerability Reports

JSON Format (vulnerabilities.json)

Machine-readable format for automated processing:

[
  {
    "id": "vuln-a3f8b2e1",
    "timestamp": "2026-03-01T12:34:56Z",
    "title": "SQL Injection in Search Endpoint",
    "severity": "critical",
    "cvss": 9.8,
    "cvss_breakdown": {
      "attack_vector": "N",
      "attack_complexity": "L",
      "privileges_required": "N",
      "user_interaction": "N",
      "scope": "U",
      "confidentiality": "H",
      "integrity": "H",
      "availability": "H"
    },
    "target": "https://example.com",
    "endpoint": "/api/search",
    "method": "GET",
    "description": "The search parameter is vulnerable to SQL injection...",
    "impact": "Complete database compromise including user credentials...",
    "technical_analysis": "The application concatenates user input directly...",
    "poc_description": "Send a GET request to /api/search with a malicious query parameter...",
    "poc_script_code": "import requests\n\npayload = \"' OR '1'='1\"\nresponse = requests.get(...)",
    "remediation_steps": "1. Use parameterized queries\n2. Implement input validation...",
    "cve": "CVE-2024-12345",
    "cwe": "CWE-89",
    "code_locations": [
      {
        "file": "src/api/search.py",
        "start_line": 42,
        "end_line": 45,
        "snippet": "query = f\"SELECT * FROM users WHERE name = '{user_input}'\"\nresults = db.execute(query)",
        "label": "Vulnerable SQL query construction",
        "fix_before": "query = f\"SELECT * FROM users WHERE name = '{user_input}'\"",
        "fix_after": "query = \"SELECT * FROM users WHERE name = ?\"\nresults = db.execute(query, (user_input,))"
      }
    ]
  }
]

Markdown Format (vulnerabilities.md)

Human-readable format for documentation and issue tracking:

# SQL Injection in Search Endpoint

**ID:** vuln-a3f8b2e1
**Severity:** CRITICAL
**Found:** 2026-03-01T12:34:56Z
**Agent:** Web Scanner
**Target:** https://example.com
**Endpoint:** /api/search
**Method:** GET
**CVE:** CVE-2024-12345
**CWE:** CWE-89
**CVSS:** 9.8
**CVSS Vector:** AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H

## Description

The search parameter is vulnerable to SQL injection, allowing attackers to 
execute arbitrary SQL commands against the database.

## Impact

Complete database compromise including extraction of user credentials, 
sensitive data, and potential for data modification or deletion.

## Technical Analysis

The application concatenates user input directly into SQL queries without 
proper sanitization or parameterization. This allows attackers to break out 
of the intended query context and execute arbitrary SQL.

## Proof of Concept

Send a GET request to /api/search with a malicious query parameter:

```python
import requests

payload = "' OR '1'='1"
response = requests.get(
    "https://example.com/api/search",
    params={"q": payload}
)
print(response.json())

Code Analysis

Location 1: src/api/search.py (lines 42-45) Vulnerable SQL query construction

query = f"SELECT * FROM users WHERE name = '{user_input}'"
results = db.execute(query)

Suggested Fix:

- query = f"SELECT * FROM users WHERE name = '{user_input}'"
+ query = "SELECT * FROM users WHERE name = ?"
+ results = db.execute(query, (user_input,))

Remediation

Use parameterized queries or prepared statements
Implement input validation and sanitization
Apply principle of least privilege to database accounts
Enable query logging and monitoring
Conduct regular security audits

## Scan Metadata

### scan_metadata.json

Contains scan configuration and summary:

```json
{
  "scan_id": "example_20260301_123456",
  "run_name": "example_20260301_123456",
  "start_time": "2026-03-01T12:34:56Z",
  "end_time": "2026-03-01T12:47:30Z",
  "duration_seconds": 754,
  "scan_mode": "deep",
  "interactive": false,
  "targets": [
    {
      "type": "web_url",
      "original": "https://example.com",
      "details": {
        "target_url": "https://example.com",
        "workspace_subdir": "target_0"
      }
    }
  ],
  "user_instructions": "Focus on authentication vulnerabilities",
  "scan_completed": true,
  "vulnerabilities_found": 3,
  "agents_spawned": 5,
  "tools_executed": 47,
  "model": "openai/gpt-4",
  "strix_version": "1.0.0"
}

Agent Traces

agent_traces.json

Detailed log of all agent activity:

{
  "agent-abc123": {
    "id": "agent-abc123",
    "name": "Web Scanner",
    "parent_id": null,
    "status": "completed",
    "start_time": "2026-03-01T12:35:00Z",
    "end_time": "2026-03-01T12:46:30Z",
    "iterations": 143,
    "messages": [
      {
        "id": "msg-1",
        "timestamp": "2026-03-01T12:35:01Z",
        "role": "assistant",
        "content": "I'll begin by mapping the application's attack surface..."
      }
    ],
    "vulnerabilities_found": 2
  }
}

Tool Executions

tool_executions.json

Complete history of tool invocations:

{
  "exec-xyz789": {
    "id": "exec-xyz789",
    "agent_id": "agent-abc123",
    "tool_name": "http_request",
    "timestamp": "2026-03-01T12:35:15Z",
    "args": {
      "url": "https://example.com/api/search",
      "method": "GET",
      "params": {"q": "test"}
    },
    "result": {
      "status_code": 200,
      "body": "..."
    },
    "duration_ms": 234,
    "status": "completed"
  }
}

Vulnerability Fields Reference

Required Fields

string

Unique identifier for the vulnerability (e.g., vuln-a3f8b2e1)

string

ISO 8601 timestamp when vulnerability was discovered

string

Brief, descriptive title of the vulnerability

string

Severity level: critical, high, medium, low, or info

number

CVSS v3.1 base score (0.0 - 10.0)

string

The target where the vulnerability was found

string

Detailed description of the vulnerability

string

Potential impact and consequences of exploitation

string

Technical details about the vulnerability

string

Description of how to reproduce the vulnerability

string

Actual exploit code or payload demonstrating the vulnerability

string

Step-by-step instructions to fix the vulnerability

CVSS Breakdown

object

CVSS v3.1 metric breakdown:

attack_vector: N (Network), A (Adjacent), L (Local), P (Physical)
attack_complexity: L (Low), H (High)
privileges_required: N (None), L (Low), H (High)
user_interaction: N (None), R (Required)
scope: U (Unchanged), C (Changed)
confidentiality: N (None), L (Low), H (High)
integrity: N (None), L (Low), H (High)
availability: N (None), L (Low), H (High)

Optional Fields

string

Specific endpoint or path where vulnerability exists

string

HTTP method (GET, POST, etc.) or operation type

string

CVE identifier if applicable (e.g., CVE-2024-12345)

string

CWE identifier (e.g., CWE-89)

array

Array of code locations relevant to the vulnerability. Each location contains:

file - Relative path to file
start_line - Starting line number
end_line - Ending line number
snippet - Code snippet showing the issue
label - Description of this location
fix_before - Original vulnerable code
fix_after - Suggested fixed code

Severity Levels

Strix uses standard severity classifications:

Severity	CVSS Range	Description
Critical	9.0 - 10.0	Immediate threat requiring urgent remediation
High	7.0 - 8.9	Significant risk requiring prompt remediation
Medium	4.0 - 6.9	Moderate risk requiring planned remediation
Low	0.1 - 3.9	Minor risk with limited impact
Info	0.0	Informational finding without direct security impact

Processing Reports

Parse JSON with jq

# Count vulnerabilities by severity
jq 'group_by(.severity) | map({severity: .[0].severity, count: length})' \
  strix_runs/*/vulnerabilities.json

# Extract all critical vulnerabilities
jq '.[] | select(.severity == "critical")' \
  strix_runs/*/vulnerabilities.json

# Get vulnerability titles and CVSS scores
jq -r '.[] | "\(.cvss) - \(.title)"' \
  strix_runs/*/vulnerabilities.json | sort -rn

# Export to CSV
jq -r '["Severity","CVSS","Title"], (.[] | [.severity, .cvss, .title]) | @csv' \
  strix_runs/*/vulnerabilities.json > vulnerabilities.csv

Parse with Python

import json
from pathlib import Path

# Load vulnerabilities
with open('strix_runs/example_20260301_123456/vulnerabilities.json') as f:
    vulns = json.load(f)

# Count by severity
from collections import Counter
severity_counts = Counter(v['severity'] for v in vulns)
print(f"Critical: {severity_counts['critical']}")
print(f"High: {severity_counts['high']}")

# Filter high-severity issues
high_severity = [v for v in vulns if v['cvss'] >= 7.0]
for vuln in high_severity:
    print(f"{vuln['cvss']:.1f} - {vuln['title']}")

# Export to GitHub Issues format
for vuln in vulns:
    title = f"[Security] {vuln['title']}"
    body = f"""
## {vuln['severity'].upper()} - CVSS {vuln['cvss']}

{vuln['description']}

### Impact
{vuln['impact']}

### Remediation
{vuln['remediation_steps']}
    """
    # Create GitHub issue via API
    # ...

Import to Jira

import json
from jira import JIRA

# Connect to Jira
jira = JIRA('https://your-instance.atlassian.net', basic_auth=('email', 'token'))

# Load vulnerabilities
with open('strix_runs/example_20260301_123456/vulnerabilities.json') as f:
    vulns = json.load(f)

# Create issues
for vuln in vulns:
    issue_dict = {
        'project': {'key': 'SEC'},
        'summary': vuln['title'],
        'description': f"""
Severity: {vuln['severity'].upper()}
CVSS Score: {vuln['cvss']}
Target: {vuln['target']}

{vuln['description']}

IMPACT:
{vuln['impact']}

REMEDIATION:
{vuln['remediation_steps']}
        """,
        'issuetype': {'name': 'Bug'},
        'priority': {'name': 'Critical' if vuln['severity'] == 'critical' else 'High'},
        'labels': ['security', 'strix', vuln['severity']]
    }
    jira.create_issue(fields=issue_dict)

Best Practices

Archive Results

Archive scan results for compliance and historical tracking:

# Create archive
tar -czf strix-scan-$(date +%Y%m%d).tar.gz strix_runs/

# Archive with metadata
tar -czf strix-scan-example-$(date +%Y%m%d).tar.gz \
  strix_runs/example_*/ \
  --transform 's|strix_runs/||'

Version Control Integration

Commit vulnerability reports for tracking:

# Copy to tracked directory
mkdir -p security-reports
cp strix_runs/latest/vulnerabilities.md security-reports/$(date +%Y-%m-%d).md
git add security-reports/
git commit -m "Security scan results $(date +%Y-%m-%d)"

Automated Reporting

Generate summary reports:

#!/bin/bash
LATEST_SCAN=$(ls -t strix_runs/ | head -1)

echo "# Security Scan Report - $(date)" > report.md
echo "" >> report.md

jq -r '.[] | "- [\(.severity | ascii_upcase)] \(.title) (CVSS \(.cvss))"' \
  "strix_runs/$LATEST_SCAN/vulnerabilities.json" >> report.md

# Email or post to Slack

Commands

Modes

Output

Output Directory Structure

Run Name Format

Vulnerability Reports

JSON Format (vulnerabilities.json)

Markdown Format (vulnerabilities.md)

Code Analysis

Remediation

Agent Traces

agent_traces.json

Tool Executions

tool_executions.json

Vulnerability Fields Reference

Required Fields

CVSS Breakdown

Optional Fields

Severity Levels

Processing Reports

Parse JSON with jq

Parse with Python

Import to Jira

Best Practices

Archive Results

Version Control Integration

Automated Reporting

See Also

​Output Directory Structure

​Run Name Format

​Vulnerability Reports

​JSON Format (vulnerabilities.json)

​Markdown Format (vulnerabilities.md)

​Code Analysis

​Remediation

​Agent Traces

​agent_traces.json

​Tool Executions

​tool_executions.json

​Vulnerability Fields Reference

​Required Fields

​CVSS Breakdown

​Optional Fields

​Severity Levels

​Processing Reports

​Parse JSON with jq

​Parse with Python

​Import to Jira

​Best Practices

​Archive Results

​Version Control Integration

​Automated Reporting

​See Also

Output Directory Structure

Run Name Format

Vulnerability Reports

JSON Format (vulnerabilities.json)

Markdown Format (vulnerabilities.md)

Code Analysis

Remediation

Agent Traces

agent_traces.json

Tool Executions

tool_executions.json

Vulnerability Fields Reference

Required Fields

CVSS Breakdown

Optional Fields

Severity Levels

Processing Reports

Parse JSON with jq

Parse with Python

Import to Jira

Best Practices

Archive Results

Version Control Integration

Automated Reporting

See Also