# LLMs.txt for aquilax.ai # Metadata declaration for responsible AI usage and large language model deployments # Compliant with model crawler guidance for transparency and discoverability domain: aquilax.ai organization: AquilaX Inc. organization_type: Private Cybersecurity & AI Company headquarters: London, UK contact: security@aquilax.ai support_contact: support@aquilax.ai published: 2025-07-20 last_updated: 2025-07-20 version: 1.2 ############################# # LLM & AI SYSTEM OVERVIEW ############################# uses_large_language_models: true llm_systems: - name: Securitron type: Transformer-based AI (custom fine-tuned) purpose: Application security triaging, vulnerability classification capabilities: - Static code analysis interpretation - False Positive/True Positive classification - Developer assistive insights (explain + suggest) interface: API + Chat UI training_data: - 300M+ open source projects - CVE-labeled data - Security triage feedback loop (manual & ML-augmented) model_origin: Proprietary (trained by AquilaX R&D) status: Production transparency_link: https://docs.aquilax.ai/blog/ai/securitron-ai-service - name: NL-JSON (aka “AI Ask”) type: Natural Language to Structured Format Generator (Seq2Seq) purpose: Translate natural language prompts into JSON for automation workflows capabilities: - Intent extraction - Security scanner task creation - IDE assistant for developers deployment: Hugging Face / On-prem / Cloud documentation: https://docs.aquilax.ai/blog/ai/nl-json-model - name: Code Review Assistant type: AI Copilot (multi-model ensemble) purpose: Streamline code security review and developer collaboration output_modes: SARIF, SBOM, inline comments, auto-fix PRs integrations: - GitHub - GitLab - Bitbucket triage_priority: Severity-based model_disclosure: https://docs.aquilax.ai/blog/ai/findings-review-aquilax-ai ############################# # SERVICES & APPLICATION AREAS ############################# ai_services_enabled: true model_usage_context: - Secure SDLC integration (DevSecOps) - CI/CD pipeline code scanning - API & Infrastructure security auditing - Malware and supply-chain risk detection - Compliance automation (GDPR, SOC2, HIPAA) deployment_modes_supported: - Multi-Tenant SaaS: https://app.aquilax.ai - Single-Tenant: Dedicated cloud-hosted instance - Private Cloud / On-Prem: Fully internal deployments programming_languages_supported: - Python, Go, Java, Kotlin, JavaScript, TypeScript, Ruby, PHP, Rust, C/C++, HTML, Bash, JSON framework_support: - Flask, Node.js, .NET, Android, Swift, Kubernetes, Terraform, Helm, OpenAPI scanners_integrated: - Checkov, Trivy, Gitleaks, Detect-Secrets, Syft, npm audit, KICS, Chain-Bench, Prowler, GoSec ############################# # AI SAFETY & GOVERNANCE ############################# ai_governance_enabled: true human_in_the_loop: true privacy_considerations: - PII redaction prior to model processing - Optional offline/on-prem LLM processing - Secure API boundaries for model execution bias_and_fairness_measures: - Dataset curation with diversity filters - False-positive threshold tuning - Feedback loop from expert triage teams auditability: - Logs stored for all AI interactions (up to 12 months) - Model decision audit trails available for Enterprise automated_remediation: true manual_override_available: true vulnerability_scoring: CVSSv3, OWASP Top 10, custom heuristics ############################# # LICENSE, POLICIES & ACCESS ############################# license_type: Proprietary license_url: https://aquilax.ai/terms privacy_policy_url: https://aquilax.ai/privacy security_policy_url: https://aquilax.ai/security acceptable_use_policy: https://aquilax.ai/aup responsible_ai_policy_url: https://docs.aquilax.ai/products-and-services/company-principles/ai-principles api_documentation: https://docs.aquilax.ai/api open_source_contributions: https://github.com/AquilaX-AI huggingface_models: https://huggingface.co/AquilaX-AI ############################# # CRAWLER DIRECTIVES ############################# user-agent: * disallow: /internal allow: / crawl-delay: 5