Technical work, shared in public.
A record of talks and workshops on MCP security, agentic AI threat modeling, prompt injection and application security, with the available presentation materials kept one click away.
About Ricardo
Ricardo Prieto is a cybersecurity practitioner focusing his career on AI safety. He has spoken at Ekoparty, UTN and community events, and shares technical work through PhiloCyber.
Languages
English and Spanish — talks, workshops and panels.
Archive
Selected talks
- 01March 2026Download deck (PDF)
ArgusTM — Multi-agent system for threat modeling
Community talk
- 02October 2025Download deck (PDF)
MCP Security for Enterprise Organizations
Ekoparty 2025 — AI Resilience Hub village
- 03October 2025Download deck (PDF)
MCP Security for Enterprise Organizations
ThreatX Security · UTN
- 04October 2024Download deck (PDF)
Prompt Injection Attacks
Community talk
- 05August 2023Download deck (PDF)
Server-Side Template Injection Workshop
Hack The Box / UTN workshop
- 06April 2023Download deck (PDF)
HTTP Request Smuggling
Community talk
- 07February 2023Download deck (PDF)
HTTP Request Smuggling (Spanish)
Community talk
Themes
Subjects in the talk record
01
MCP Security
Securing Model Context Protocol deployments: tool poisoning, authentication, authorization and enterprise rollout patterns.
02
Agentic AI Threat Modeling
Threat modeling for multi-agent systems — from the A2AS framework to hands-on methodology your team can apply.
03
Prompt Injection
Direct and indirect prompt injection attacks against LLM applications, and the defenses that actually hold up.
04
AI Security Governance
Policy, risk and secure adoption of AI in the enterprise — governance that enables instead of blocking.
Speaking / Public record
Have a relevant event or collaboration in mind?
Send the audience, format, timing and why the subject is a useful fit. I welcome specific invitations connected to AI safety, AI security and the work documented here.