Overview
News
Technologies
Salaries
Products
People
Growth
Financials
Overview
The AI firewall between your app and OpenAI, Anthropic, Mistral or Gemini — plus a verifiable, tamper-evident audit trail that proves what your AI did with your data. EU-hosted, self-hostable (incl. Azure), EU AI Act record-keeping (Article 12 & 19).
News
They'll Verify. They Just Won't Act. - Senthex Research
A single external issue induced a five-agent CI/CD pipeline to deploy secret-exfiltrating code in up to 55% of worst-case runs. A competent LLM scanner passed 80% of laundered pull requests; zero content-based controls caught it. 100% synthetic, fully rep
Read more
Report
A prompt injection nearly hijacked my coding agent mid-task - Senthex Blog
Last week a piece of tool output impersonated me and nearly redirected my coding agent to a task I never asked for. A first-hand look at indirect prompt injection - and the trust boundary that has to be engineered in.
Read more
Report
Solutions - AI security & verifiable audit, by the problem you're solving
Answers to what you're actually trying to do: prove what your AI did, reconstruct a decision after an incident, detect prompt injection, secure AI agents, and keep your AI logs in the EU. EU-hosted, self-hostable, honest about what's shipping vs the v1.1.
Read more
Report
Senthex - AI Firewall + Verifiable Audit Trail for LLMs
The AI firewall between your app and OpenAI, Anthropic, Mistral or Gemini - plus a verifiable, tamper-evident audit trail that proves what your AI did with your data. EU-hosted, self-hostable (incl. Azure), EU AI Act record-keeping (Article 12 & 19).
Read more
Report