Security Audit
Deep audits with prioritized fixes, SBOM hygiene checks, and exportable reports (JSON/PDF). Ideal for quick wins and executive summaries.
Built by stm.ai and funded by a €1.7M non-dilutive grant, CLARA combines a security-specialized LLM (CLARA Core) with deployable agents for security audits, network monitoring, and secure browsing. Hybrid deploy. Explainable by design.
Security-specialized model that orchestrates agents and produces human-readable rationales.
A security-specialized core model with three production-ready agents, coordinated through a unified admin dashboard and an education platform that turns feedback into better agents.
Deep audits with prioritized fixes, SBOM hygiene checks, and exportable reports (JSON/PDF). Ideal for quick wins and executive summaries.
Real-time anomaly detection and alerting with human-readable explanations and suppression of noisy patterns to reduce alert fatigue.
In-session phishing and malware defense, URL risk scoring, and safe-action suggestions for employees without breaking workflows.
Deploy in hours. Start with Audit + Browser Protection. See tangible fixes and compliance uplift in week one.
Data locality, on-prem options, and explainable decisions. Align with GDPR, AI Act, and CRA from day one.
Education Platform to train and certify staff. Turn feedback into better agents using privacy-preserving learning.
This site is all about CLARA. You’re in the right place.
Our LLM education/coaching product with a standalone site.
Go to llmentor.com →Preview only. Final pricing and packaging will ship at launch. Contact us for the latest.
Preview only. Full documentation and SDKs are launching soon.
Org key created Target connected POST /v1/agents/audit/run GET /v1/reports/{id} POST /v1/agents/audit/run GET /v1/reports/{id} POST /v1/alerts/ack No raw log exfiltration by default. Keys via KMS. Federated learning + differential privacy options. Full auditability.
stm.ai is a product company focused on building CLARA, our AI cybersecurity platform. We also created LLMentor.
No raw log exfiltration by default. Keys via KMS. Federated learning + differential privacy options. Full auditability.
On-prem, cloud, or air-gapped deployment options.
Human-readable rationales with every recommendation.
Privacy-preserving learning and minimal data capture.
Want early access, a demo, or pricing details? Email us at c@stm.ai and tell us a bit about your environment and goals.
We collect only what you send us by email. Purpose: to contact you about CLARA (product information, demos, onboarding). Lawful basis: consent. Data controller: AI STM Learning SRL.
AI STM Learning SRL, as Beneficiary of Contract no. 390016/28.08.2025, announces the launch of the project “CLARA (CyberSecurity Learning And Response Agent) - Innovative Cybersecurity Solution”, SMIS code 334596, submitted under the project call PCIDIF/155/PCIDIF_P1/OP1/RSO1.1/PCIDIF_A1 - Support for newly established innovative enterprises.
The general objective of the project is to increase the innovation capacity and competitiveness of AI STM Learning SRL in cybersecurity by developing, implementing, and launching on the market an advanced software solution that combines VPN technology with artificial intelligence, providing companies with dynamic and proactive protection against continuously evolving cyber threats.
The total project value is 7.659.236,30 RON, of which ERDF value: 6.443.369,15 RON and eligible non-reimbursable value from the national budget: 1.137,065,15 RON.
Project implementation period: 28.08.2025 - 27.08.2027.
Project co-financed by the European Regional Development Fund (ERDF), through the Smart Growth, Digitalisation and Financial Instruments Programme 2021-2027.