| user: | lbeurerkellner |
| created: | April 5, 2023 |
| karma: | 203 |
| 1. | |
| 2. | |
| 3. | |
| 4. | |
| 5. | |
| 6. | Exposing Novel Prompt Injections, "Toxic Flows" in Agentic Systems(invariantlabs.ai) |
| 7. | Hardening AI Agent Security with Toxic Flow Analysis(invariantlabs.ai) |
| 8. | |
| 9. | |
| 10. | |
| 11. | Show HN: Guardrails – A Contextual Security Layer for Agentic AI Systems(invariantlabs.ai) |
| 12. | Securing Agents with Rules(explorer.invariantlabs.ai) |
| 13. | 1 year ago | discuss |
| 14. | |
| 15. | AgentDojo: An Environment to Evaluate Prompt Injection Attacks on AI Agents(agentdojo.spylab.ai) |
| 16. | WhatsApp MCP Exploited: Exfiltrating your message history via MCP(invariantlabs.ai) |