For AI models & agents · EU AI Act · NIST RMF · NYC LL144

Stress-test your AI models and agents
the way real attackers will

Aegis runs adversarial attacks against your AI — from a plain chatbot to a tool-using agent — the way real attackers will: multi-turn social engineering, tool-call abuse, indirect injection. Then it produces the audit-ready evidence regulators, auditors, and your CISO actually demand.

No credit card to run your first assessment.

Most AI security tools test single prompts. Aegis tests the whole system.

Single-prompt scanners catch obvious jailbreaks. They miss the failure modes that actually break AI in production — whether it’s a plain chatbot or a tool-using agent: a 3-turn rapport-building con that flips an HR-bot into giving salary data, a roleplay that talks a banking agent into a tool-call that moves money, a poisoned doc that hijacks a RAG pipeline.

Aegis treats your AI the way attackers do — and ships the proof.

Multi-turn attacks

Scripted attacker conversations that build rapport, launder context, and pivot — not just one-shot prompt injection. Agents fail differently in conversation than they do in isolation.

Tool-call inspection

Declare your agent’s tools (send_email, transfer_funds, exec_code). We capture every call it emits under attack and score destructive actions as failures even when the text reply is polite.

3-layer escalation

Baseline → authority/urgency reframe → multi-layered social engineering. Each pass adapts to what the previous one revealed, so we find what one-shot scanners miss.

Built for the audit, not the dashboard

Every assessment produces one report: a 30-second executive briefing for the C-suite, severity-rated findings with evidence, a regulator-mapped compliance gap analysis, and a prioritized remediation plan. Shareable to legal with one link.

Audit-ready

EU AI Act / NIST RMF / GDPR / NYC LL144 mapping per finding.

Restricted sharing

Email-allowlist link with OTP verification for sensitive reports.

Full evidence appendix

Every attack run, every transcript, every tool call — preserved for auditors.

6 policy packs, regulator-mapped

Every finding cites the specific regulation it implicates.

Prompt Injection
PII Leakage
Harmful Content
Unauthorized Tool Use
Regulatory Control
Bias & Fairness
EU AI Act Art.9/10/15NIST AI RMFNIST AI Bias SP 1270GDPROWASP LLM Top 10NYC Local Law 144EEOC AI Guidance

From endpoint to audit report in minutes

1

Connect your AI

Point us at any OpenAI-compatible endpoint — a chatbot, a RAG app, or a tool-using agent (LangChain, LangServe, Open WebUI, OpenAI Assistants, or your own wrapper). Declare the tools your agent exposes if you want tool-call inspection.

2

Run an assessment

Pick policy packs. Aegis runs single-turn AND multi-turn attacks, escalating up to 3 passes per template. Captures everything: text replies and tool invocations.

3

Share the audit report

Executive briefing, compliance gaps, full evidence appendix with every conversation transcript. Send the link to legal or restrict to specific emails. Print to PDF for the board pack.

Simple, transparent pricing

First assessment is free. Cancel any plan anytime, no refunds.

Free

$0forever

Probe one AI system with a baseline attack pack and see the audit-ready report.

  • 1 assessment per month
  • Prompt Injection pack (10 attack vectors)
  • Tests any OpenAI-compatible chat or agent endpoint
  • Audit-ready report — public share link
  • No credit card required

Starter

$299per month

For teams shipping AI assistants and agents that need regular adversarial coverage.

  • 5 assessments per month
  • 3 policy packs (Prompt Injection, PII, Harmful Content)
  • Multi-turn attack scenarios (agent-aware)
  • Audit-ready PDF reports
  • Public share links
  • Email support
Most popular

Enterprise

$999per month

For regulated teams shipping agents that take real-world actions.

  • 100 assessments per month
  • All 6 policy packs (incl. Bias & Fairness)
  • Tool-call inspection — flag dangerous agent actions
  • Full evidence appendix (every attack run)
  • Restricted share links — email allowlist + OTP verification
  • Compliance mapping: EU AI Act, NIST, GDPR, NYC LL144
  • Priority support · 24h SLA

Ship your AI. Prove it’s safe.

Run your first adversarial assessment free. See exactly how your AI fails under multi-turn pressure, tool-call abuse, and the failure modes single-prompt scanners miss.