Appearance
Use the OWASP Top 10 for LLM Applications in Modulos
Modulos has one Framework for the OWASP Top 10 for LLM Applications: MFF-11, with the name OWASP - Top 10 for LLM Applications. You add MFF-11 to the AI System of an LLM application. Modulos has no Framework for the OWASP Top 10 for LLM Applications at the Organization level.
The OWASP Top 10 for LLM Applications is a security guideline. It is not law, and it gives no obligations. See Use of the OWASP Top 10.
The OWASP Top 10 in the Framework
The Modulos Framework MFF-11 maps the OWASP Top 10 for LLM Applications 2025, with the risks LLM01:2025 to LLM10:2025.
For the changes in the OWASP Top 10 for LLM Applications 2026, see Changes in 2025 and 2026.
Add the Framework
For the OWASP Top 10 for LLM Applications, add MFF-11 to the AI System of each LLM application.
- Make an AI System for the LLM application in the AI Registry, or open its AI System.
- For a new AI System, select
MFF-11when you make the AI System. - For an AI System that you made before, add
MFF-11to it.
For the procedure, see "Register an AI System" and Framework versions.
For a model that starts "acting on its own", you can use the OWASP Top 10 for Agentic Applications Framework with MFF-11 (Use of the OWASP Top 10).
The Requirements and the OWASP risks
For each MFF-11 Requirement, select the OWASP risk below to read the measures that OWASP recommends.
| Requirement | Requirement name | OWASP 2025 risk |
|---|---|---|
MRF-223 | LLM01:2025 - Prompt Injection | LLM01:2025 Prompt Injection |
MRF-224 | LLM02:2025 - Sensitive Information Disclosure | LLM02:2025 Sensitive Information Disclosure |
MRF-225 | LLM03:2025 - Supply Chain | LLM03:2025 Supply Chain |
MRF-226 | LLM04:2025 - Data and Model Poisoning | LLM04:2025 Data and Model Poisoning |
MRF-227 | LLM05:2025 - Improper Output Handling | LLM05:2025 Improper Output Handling |
MRF-228 | LLM06:2025 - Excessive Agency | LLM06:2025 Excessive Agency |
MRF-229 | LLM07:2025 - System Prompt Leakage | LLM07:2025 System Prompt Leakage |
MRF-230 | LLM08:2025 - Vector and Embedding Weaknesses | LLM08:2025 Vector and Embedding Weaknesses |
MRF-231 | LLM09:2025 - Misinformation | LLM09:2025 Misinformation |
MRF-232 | LLM10:2025 - Unbounded Consumption | LLM10:2025 Unbounded Consumption |
For the Controls that are in MFF-11 and in other Frameworks, see Control overlap.
Do the work
For the OWASP Top 10 for LLM Applications Framework, do these steps for each Requirement.
- Read the OWASP risk in The Requirements and the OWASP risks.
- Do the work of each Control of the Requirement.
- Attach each Evidence item to its Control.
- When the work of a Control is complete, set the status of the Control.
For Controls, see Controls.
Evidence and test results are not the security measures that OWASP recommends (see Use of the OWASP Top 10).
MFF-11 has Requirements and Controls. It has no tests and no schedules. If you use tests in Modulos, do these steps:
- Prepare the tests that are necessary for the LLM application.
- Attach each test to its Controls.
- To do a test at intervals, give it a schedule.
For tests and schedules, see Tests and schedules.
Review the Requirements
In Modulos, the Requirement Owner does the Review of each Requirement of the OWASP Top 10 for LLM Applications Framework. The Review can start when each Control of the Requirement has a final status. The Requirement Owner examines the completed work. Then the Requirement Owner marks the Requirement as Fulfilled.
Controls with a final status, Evidence, and test results do not make a Requirement Fulfilled. For the conditions of a Review, see Requirements and Reviews and statuses.
Related information
- OWASP Top 10 for LLM Applications: the 2025 risks, the 2026 changes, and other Frameworks.
- The FAQ of the OWASP Top 10 for LLM Applications.
- "Register an AI System": how to make an AI System in Modulos.
- Requirements: Requirements, their status, and their Review.
- Control overlap: the Controls that more than one Modulos Framework uses.
Sources
- OWASP GenAI Security Project, OWASP Top 10 for LLM Applications 2025, Version 2025, November 18, 2024: the risks LLM01:2025 to LLM10:2025.
- OWASP GenAI Security Project, OWASP Top 10 for LLM Applications 2026, with the "Release date" August 4, 2026: the Letter from the Project Leads, "What's New in the 2026 Top 10".
- The Modulos Framework
MFF-11and its RequirementsMRF-223toMRF-232.
Disclaimer
This page gives general information. It is not legal advice.