Documentation
Introduction
Trakory is an AI-powered code security auditing platform built by Nexoid Ltd. It scans your repositories for security vulnerabilities, hardcoded secrets, misconfigurations, and OWASP Top 10 issues. Trakory provides automated analysis, AI-assisted remediation, and team-wide visibility into your code security posture.
Every line of code. Every vulnerability scanned. In one platform.
Supported Languages
What Trakory Detects
- Security vulnerabilities (SQL injection, XSS, command injection, path traversal)
- Hardcoded secrets and credentials (API keys, tokens, passwords, private keys)
- Security misconfigurations (CORS, CSP, authentication, rate limiting)
- OWASP Top 10 vulnerabilities
- Data flow and taint analysis
- Dependency vulnerabilities
Getting Started
Creating an Account
Visit the signup page to create a Trakory account. You can sign up with email and password, or use GitHub OAuth for a faster setup. Free accounts are limited to 10 scans per month and public repositories only.
Connecting a Repository
Once logged in, use the dashboard chat to scan a repository. You can scan a GitHub repository by pasting its URL or by connecting your GitHub account. The dashboard supports the following commands:
/github <url>Scan a GitHub repository by URL/website <url>Analyze a public website/search <query>Search through past scan resultsRunning Your First Scan
- Navigate to the dashboard.
- Click “New Scan” in the sidebar or type a repository URL in the chat.
- Select a repository from the GitHub panel or paste a URL directly.
- Wait for the scan to complete. Results appear in real-time with severity ratings.
- Click on any finding to see detailed information and remediation suggestions.
Dashboard Guide
The dashboard is the central hub for all scanning activity. It consists of several panels:
Sidebar
The left sidebar provides access to all authenticated pages: Dashboard, History, Repositories, Settings, and Admin (if you have admin privileges). It also shows your current credit balance and account information.
Chat Interface
The main area features an AI-powered chat interface. Send messages to scan repositories, ask questions about results, or get remediation guidance. The chat supports markdown rendering for formatted findings and code snippets.
Session History
Accessible from the left panel, the session history shows all past scans with repository names, timestamps, issue counts, and status indicators. You can delete old sessions or click to revisit results.
Activity Panel
The right panel provides a real-time activity feed showing scan completions, new findings, auto-fixes, PR creation, and file modifications. Each activity type has a distinct color and icon for quick identification.
Provider Selector
Choose which AI model powers your scans. Options include free built-in models and your own API keys for BYO (bring your own) providers. You can configure custom API keys in Settings.
Scanning
Scan Lifecycle
Each scan goes through the following stages:
Severity Levels
Supported Analysis Types
- Static Analysis: AST-based pattern matching for code vulnerabilities.
- Secret Detection: Regex and entropy-based detection of hardcoded credentials.
- Dependency Scan: Checks for known vulnerabilities in dependencies.
- Data Flow Analysis: Tracks untrusted data through your application.
- Configuration Review: Checks security configs, CSP, CORS, authentication.
- OWASP Top 10: Full coverage of the OWASP Top 10 vulnerability categories.
AI Remediation
Trakory uses AI to not just detect vulnerabilities, but also help fix them. On Pro and Enterprise plans, you can:
- Auto-Fix: Generate secure code replacements for detected vulnerabilities.
- PR Creation: Automatically create a pull request with the fix applied.
- Contextual Guidance: Get detailed remediation advice specific to your codebase.
- Provider Flexibility: Use built-in AI models or bring your own API key.
Bring Your Own AI Model
Configure custom API keys in Settings to use your preferred AI provider. Supported providers include OpenAI, Anthropic, Google, and any OpenAI-compatible endpoint. Each key can be assigned a specific model and tested before use.
CI/CD Integration
Integrate Trakory into your CI/CD pipeline to automatically scan every push. Add the following GitHub Action to your workflow:
- name: Trakory Security Scan
uses: trakory/security-scan@v1
with:
api-key: ${{ secrets.TRAKORY_API_KEY }}Scans run automatically on every push. Results appear as PR comments and check annotations. You can configure the action to block PRs if critical or high-severity issues are found.
Get your API key from the Settings page once you have an active subscription.
Custom Rules
Define your own security patterns using the Trakory rule format. Custom rules allow you to enforce organization-specific security policies and detect internal patterns.
Rule Format
Rules are YAML-based and support regex matching, AST patterns, and data-flow analysis. Store custom rules in a .trakory/ directory at your repository root.
rule: id: "custom-rule-001" name: "No Debug Endpoints" severity: high pattern: "app\.(get|post)\(.*debug.*\)" description: "Debug endpoints should not be present in production code." category: "security"
Enterprise plans include support for custom rule creation, rule testing, and organization-wide rule distribution.
BYO API Keys
If you prefer to use your own AI provider or need access to specific models, you can configure custom API keys in the Settings page. Trakory supports:
- OpenAI (GPT-4, GPT-4o, o3, and any compatible model)
- Anthropic (Claude 3.5 Sonnet, Claude 4 Opus)
- Google (Gemini 2.5 Pro, Gemini 2.5 Flash)
- Any OpenAI-compatible endpoint (together.ai, fireworks.ai, deepseek, etc.)
Each API key configuration includes a provider URL, API key, model selection, and a test button to verify connectivity. Keys are stored securely and used only for your scans.
Notifications
Trakory provides real-time notifications to keep you informed about scan results and platform events.
Notification Types
- Scan Complete: A repository scan has finished processing.
- Vulnerability Found: A new security issue has been detected.
- Credits Low: Your scan credit balance is running low.
- System: Platform announcements and maintenance notices.
- Scan Progress: Real-time updates during an active scan.
Managing Notifications
Configure your notification preferences in the Settings page. You can enable or disable each notification type independently. The notification bell in the sidebar shows unread notifications with a badge count.
Billing & Plans
Trakory offers three subscription tiers to match your needs. Visit the Billing page to manage your subscription.
Free Plan
$0 per month — 10 scans per month (resets weekly), public repositories only, basic vulnerability detection, email support.
Pro Plan
$29 per month — Unlimited scans, private repositories, OWASP Top 10+ coverage, AI remediation with auto-fix and PR creation, priority support.
Enterprise Plan
Custom pricing — Everything in Pro plus on-premise deployment, custom rules engine, SLA guarantees, dedicated support team, organization-wide policy management.
Billing is handled securely through Stripe. You can upgrade, downgrade, or cancel your subscription at any time. All plans include a 14-day free trial for Pro features.
Security & How We Handle Your Code
We take the security of your code seriously. Here is how Trakory protects your repositories and data:
Sandboxed Scanning
Repositories are cloned to sandboxed, isolated temporary directories. Each scan runs in its own ephemeral environment that is destroyed immediately after the scan completes.
No Permanent Storage
Your source code is never stored permanently. It is cloned only for the duration of the scan, analyzed in memory, and the temporary directory is cleaned automatically.
Encryption
All data in transit is encrypted using TLS 1.2+. Data at rest is encrypted using AES-256. API keys and secrets are stored securely.
Access Controls
Strict access controls and authentication requirements are enforced for all personnel. Multi-factor authentication is supported via GitHub OAuth.
Security Headers
The platform implements strict security headers including Content Security Policy, HTTP Strict Transport Security, X-Frame-Options, X-Content-Type-Options, and Permissions Policy.
Vulnerability Disclosure
If you discover a security issue in Trakory itself, please report it to contact@nexoid.com. We practice responsible disclosure and will respond promptly.
Admin Panel
The admin panel is available to users with administrator privileges. It provides system-wide visibility and management tools.
Dashboard
Overview cards showing total users, total scans completed, error counts, and active sessions. Key metrics at a glance for monitoring platform health.
Error Logs
Searchable, paginated error log viewer with detailed stack traces and user context. Filter by error type, severity, and date range.
Session Monitoring
View all active and historical scan sessions. Monitor scan progress, resource usage, and completion status across all users.
User Management
View all registered users with account details, subscription tier, scan counts, and account status. Manage user roles and access levels.
Frequently Asked Questions
What types of vulnerabilities does Trakory detect?
Trakory detects a wide range of security issues including SQL injection, cross-site scripting (XSS), hardcoded secrets, security misconfigurations, OWASP Top 10 vulnerabilities, dependency vulnerabilities, and data flow issues.
How fast is a typical scan?
Scan speed depends on repository size. Small repositories scan in under a minute. Large monorepos may take several minutes. The scan provides real-time progress updates throughout the process.
What does the free tier include?
The free tier includes 10 scans per month (resets weekly), public repository support, and basic vulnerability detection. No credit card is required to start.
Which languages are supported?
Trakory currently supports Python, JavaScript, TypeScript, Go, Rust, and Java. Additional language support is being added regularly.
Can Trakory automatically fix vulnerabilities?
Yes. Pro and Enterprise plans include AI-powered auto-fix capabilities that generate secure code replacements and can create pull requests with the fixes applied.
Is my code safe with Trakory?
Yes. Your code is cloned to a sandboxed temporary directory, analyzed in memory, and automatically cleaned up after the scan completes. Code is never stored permanently.
Can I use my own AI model?
Yes. The Settings page lets you configure custom API keys for OpenAI, Anthropic, Google, or any OpenAI-compatible endpoint.
How do I cancel my subscription?
You can cancel your subscription at any time from the Billing page. Access to Pro features continues until the end of the current billing period.
Do you offer on-premise deployment?
Yes. Enterprise plans include on-premise deployment options. Contact us at contact@nexoid.com for more information.
Support
Need help? Trakory offers multiple support channels:
- Email Support: contact@nexoid.com — For all inquiries, including billing, technical issues, and general questions.
- Support Portal: Submit a ticket — Directly from the platform with category selection (bug report, feature request, account issue, billing question, other).
- Priority Support: Pro and Enterprise plans include priority support with faster response times and dedicated account management for Enterprise customers.
Enterprise customers receive a dedicated support team with guaranteed SLA response times. Contact us for more information about Enterprise plans and custom support arrangements.
Nexoid Ltd. | https://www.trakory.com | contact@nexoid.com