> ## Content Index
> Fetch the complete content index at: https://www.humanoidthreats.com/llms.txt
> Use this file to discover other available public pages before exploring further.

# OWASP GenAI Security Project
- URL: https://www.humanoidthreats.com/directory/owasp-genai-security-project/
- Published: 2026-11-01T07:26:09.000Z
- Updated: 2026-11-01T07:26:09.000Z
- Description: OWASP flagship project behind the Top 10 for LLM Applications and the Top 10 for Agentic Applications, publishing free GenAI security guidance.
- Author: Theo Marchetti
- Tags: Directory, #dir-open

Category

Open-source AI security guidance

Website

[genai.owasp.org](https://genai.owasp.org/?ref=humanoidthreats.com)

Last verified

November 1, 2026

## What is OWASP GenAI Security Project?

The OWASP GenAI Security Project is a global, community-driven initiative for identifying and tackling the risks of generative AI systems and applications. It is an OWASP Foundation Flagship Project and grew out of the OWASP Top 10 for LLM Applications. It publishes free, open guidance, including the 2026 edition of the Top 10 for LLM Applications released on September 1, 2026, and the OWASP Top 10 for Agentic Applications released in December 2025.

## Key capabilities

- **Top 10 for LLM Applications,** with the 2026 edition mapped to NIST, MITRE ATLAS and CWE.
- **Top 10 for Agentic Applications,** covering risks from agent goal hijack and tool misuse to rogue agents.
- **Agent Control Standard** for runtime agent enforcement.
- **GenAI Red Teaming Manual** and exploit roundups.
- **Initiatives** on agentic app security and AI security governance.

## Why it matters for humanoid robot security

Editorial analysis: the project does not address robots directly, but its agentic list reads like a threat model for an AI-driven humanoid. Agent goal hijack, tool misuse, memory and context poisoning, cascading failures and human-agent trust exploitation all have physical counterparts when the agent controls a body. The 2026 LLM list ranks excessive agency third, which is the core concern when a model can move objects or open doors.

## Where it fits in a humanoid robot security program

Analysis: OWASP's lists fit at the AI model and agent layer, and at the design and procurement stages. Robot teams can use them as a requirements checklist for the planner and assistant, and buyers can ask vendors to show how each risk is handled. The red teaming manual helps structure pre-release testing, and the Agent Control Standard points toward runtime controls on what an agent may do.

## Who it is for

- Robot software teams building LLM or agent features
- Buyers who need a vendor-neutral checklist for AI risks
- Researchers mapping physical AI risks to established categories

## Questions to ask OWASP GenAI Security Project

- How do the agentic Top 10 risks change when an agent controls actuators?
- Is there work underway on embodied or physical AI guidance?
- How does the Agent Control Standard handle real-time constraints?
- How can robotics practitioners contribute to the project?

## Similar listings

- [MITRE ATLAS](https://www.humanoidthreats.com/directory/mitre-atlas/): AI threat framework
- [garak (NVIDIA)](https://www.humanoidthreats.com/directory/garak-nvidia/): Open-source LLM scanner
- [SROS2 (ROS 2 security)](https://www.humanoidthreats.com/directory/sros2/): Open-source robot security

## Related on Humanoid Threats

- [Can LLM-controlled robots be jailbroken?](https://www.humanoidthreats.com/can-llm-controlled-robots-be-jailbroken/)
- [Prompt injection against robot VLA models](https://www.humanoidthreats.com/prompt-injection-robot-vla-models/)
- [Browse the full Humanoid Robot Security Directory](https://www.humanoidthreats.com/directory/)

## Sources

- [OWASP GenAI Security Project website](https://genai.owasp.org/?ref=humanoidthreats.com)
- [OWASP Foundation project page](https://owasp.org/www-project-genai-security/?ref=humanoidthreats.com)
- [OWASP Top 10 for Agentic Applications](https://genai.owasp.org/2025/12/09/owasp-top-10-for-agentic-applications-the-benchmark-for-agentic-security-in-the-age-of-autonomous-ai/?ref=humanoidthreats.com)
- [OWASP: 2026 Top 10 for LLM Applications announcement](https://genai.owasp.org/2026/09/01/owasp-genai-security-project-unveils-2026-top-10-for-llm-applications-new-agent-control-standard-and-sponsors-as-community-tops-30000-members/?ref=humanoidthreats.com)

*This listing is editorial and free; inclusion is not an endorsement. Work at OWASP GenAI Security Project?* [*Claim or update this listing*](mailto:info@humanoidthreats.com?subject=Update%20listing%3A%20owasp-genai-security-project)*.*

Get the weekly Humanoid Threats Brief → [Subscribe](#/portal/signup)