DeepTeam is a framework to red team LLMs and AI agents.
updated docs
penguined79f993fedb1
diff --git a/docs/docs/frameworks-introduction.mdx b/docs/docs/frameworks-introduction.mdxindex 68ecd275..65286b57 100644--- a/docs/docs/frameworks-introduction.mdx+++ b/docs/docs/frameworks-introduction.mdx@@ -4,8 +4,8 @@ title: Introductionsidebar_label: Introduction----The DeepTeam Frameworks define structured methodologies for **AI red teaming and risk assessment**.-Each framework maps to a recognized safety or security standard, helping you test your model's robustness against **real-world adversarial behavior, dataset risks, and system vulnerabilities**.+The DeepTeam Frameworks define structured methodologies for **AI red teaming and risk assessment**.+Each framework maps to a recognized safety or security standard, helping you test your model's robustness against **real-world adversarial behavior, dataset risks, and system vulnerabilities**.DeepTeam supports multiple frameworks — from dataset-based testing (_BeaverTails, Aegis_) to security and governance standards (_MITRE ATLAS, NIST AI RMF, OWASP Top 10 for LLMs_).@@ -13,13 +13,38 @@ DeepTeam supports multiple frameworks — from dataset-based testing (_BeaverTaiHere are the list of frameworks available in `deepteam`:-- [OWASP Top 10 for LLMs](#owasp-top-10-for-llms)+- [OWASP](#owasp-top-10)+ - OWASP Top 10 for Agents 2026+ - OWASP Top 10 for LLMs 2025- [NIST AI RMF](#nist-ai-rmf)- [MITRE ATLAS](#mitre-atlas)- [BeaverTails](#beavertails)- [Aegis](#aegis)-### OWASP Top 10 for LLMs+### OWASP Top 10++#### OWASP Top 10 for Agents 2026++The OWASP Top 10 for Agentic Applications (ASI) identifies the most critical security risks introduced by autonomous and semi-autonomous AI agents.+It focuses on failures arising from goal misalignment, tool misuse, delegated trust, inter-agent communication, persistent memory, and emergent autonomous behavior.++- Tests for goal hijacking, tool misuse, identity abuse, memory poisoning, and rogue agents+- Simulates cascading failures and inter-agent communication vulnerabilities+- Ideal for multi-agent systems, tool-using agents, and autonomous AI applications++```python+from deepteam.frameworks import OWASP_ASI_2026+owasp_asi = OWASP_ASI_2026(num_attacks=10)++risk = red_team(+ model_callback=your_model_callback,+ framework=owasp_asi+)+```++[Learn more about OWASP Top 10 for Agents 2026](/docs/frameworks-owasp-top-10-for-agentic-applications)++#### OWASP Top 10 for LLMs 2025The OWASP Top 10 for LLMs framework identifies the most critical security risks in LLM applications.It reflects the 2025 OWASP edition, covering vulnerabilities in RAG systems, agents, and model integrations.@@ -33,7 +58,7 @@ from deepteam.frameworks import OWASPTop10owasp = OWASPTop10(num_attacks=10)risk = red_team(- model_callback=your_model_callback,+ model_callback=your_model_callback,framework=owasp)```@@ -53,7 +78,7 @@ from deepteam.frameworks import NISTnist = NIST(num_attacks=10)risk = red_team(- model_callback=your_model_callback,+ model_callback=your_model_callback,framework=nist)```@@ -65,9 +90,9 @@ risk = red_team(The MITRE ATLAS framework integrates the MITRE ATLAS knowledge base, focusing on adversarial tactics and techniques used against AI systems.It evaluates system resilience across attack phases like Reconnaissance, Resource Development, Initial Access, and Impact.-- Tests adversarial behavior patterns from the ATLAS taxonomy-- Detects vulnerabilities such as prompt injection, data poisoning, and exfiltration-- Ideal for AI security simulation and penetration testing+- Tests adversarial behavior patterns from the ATLAS taxonomy+- Detects vulnerabilities such as prompt injection, data poisoning, and exfiltration+- Ideal for AI security simulation and penetration testing```pythonfrom deepteam.frameworks import MITREATLAS@@ -77,7 +102,7 @@ from somewhere import your_model_callbackatlas = MITREATLAS(num_attacks=10)risk = red_team(- model_callback=your_model_callback,+ model_callback=your_model_callback,framework=atlas)```@@ -97,7 +122,7 @@ from deepteam.frameworks import BeaverTailsbeaver = BeaverTails(num_attacks=10)risk = red_team(- model_callback=your_model_callback,+ model_callback=your_model_callback,framework=beaver)```@@ -117,7 +142,7 @@ from deepteam.frameworks import Aegisaegis = Aegis(num_attacks=10)risk = red_team(- model_callback=your_model_callback,+ model_callback=your_model_callback,framework=aegis)```@@ -126,4 +151,4 @@ risk = red_team(:::tipYou can customize and add more attacks and vulnerabilities to already existing frameworks to specialise red-teaming to your LLM's use case.-:::\ No newline at end of file+:::diff --git a/docs/docs/frameworks-owasp-top-10-for-agentic-applications.mdx b/docs/docs/frameworks-owasp-top-10-for-agentic-applications.mdxindex 50562d80..9072f187 100644--- a/docs/docs/frameworks-owasp-top-10-for-agentic-applications.mdx+++ b/docs/docs/frameworks-owasp-top-10-for-agentic-applications.mdx@@ -1,7 +1,7 @@---id: frameworks-owasp-top-10-for-agentic-applications-title: OWASP Top 10 for Agents-sidebar_label: OWASP Top 10 for Agents+title: OWASP Top 10 for Agents 2026+sidebar_label: OWASP Top 10 for Agents 2026---The **OWASP Top 10 for Agentic Applications (ASI)** identifies the most critical security risks introduced by autonomous and semi-autonomous AI agents. Unlike traditional LLM applications, agentic systems combine reasoning, memory, tools, and multi-step execution, introducing new classes of vulnerabilities that extend beyond prompt-level attacks.@@ -76,8 +76,8 @@ vulnerabilities = owasp_asi.vulnerabilities# Modify attributes for your specific testing context if neededred_team(- model_callback=your_model_callback,- attacks=attacks,+ model_callback=your_model_callback,+ attacks=attacks,vulnerabilities=vulnerabilities)```@@ -108,8 +108,8 @@ vulnerabilities = owasp_asi.vulnerabilities# Modify attributes for your specific testing context if neededred_team(- model_callback=your_model_callback,- attacks=attacks,+ model_callback=your_model_callback,+ attacks=attacks,vulnerabilities=vulnerabilities)```@@ -140,8 +140,8 @@ vulnerabilities = owasp_asi.vulnerabilities# Modify attributes for your specific testing context if neededred_team(- model_callback=your_model_callback,- attacks=attacks,+ model_callback=your_model_callback,+ attacks=attacks,vulnerabilities=vulnerabilities)```@@ -172,8 +172,8 @@ vulnerabilities = owasp_asi.vulnerabilities# Modify attributes for your specific testing context if neededred_team(- model_callback=your_model_callback,- attacks=attacks,+ model_callback=your_model_callback,+ attacks=attacks,vulnerabilities=vulnerabilities)```@@ -204,8 +204,8 @@ vulnerabilities = owasp_asi.vulnerabilities# Modify attributes for your specific testing context if neededred_team(- model_callback=your_model_callback,- attacks=attacks,+ model_callback=your_model_callback,+ attacks=attacks,vulnerabilities=vulnerabilities)```@@ -236,8 +236,8 @@ vulnerabilities = owasp_asi.vulnerabilities# Modify attributes for your specific testing context if neededred_team(- model_callback=your_model_callback,- attacks=attacks,+ model_callback=your_model_callback,+ attacks=attacks,vulnerabilities=vulnerabilities)```@@ -267,8 +267,8 @@ vulnerabilities = owasp_asi.vulnerabilities# Modify attributes for your specific testing context if neededred_team(- model_callback=your_model_callback,- attacks=attacks,+ model_callback=your_model_callback,+ attacks=attacks,vulnerabilities=vulnerabilities)```@@ -299,8 +299,8 @@ vulnerabilities = owasp_asi.vulnerabilities# Modify attributes for your specific testing context if neededred_team(- model_callback=your_model_callback,- attacks=attacks,+ model_callback=your_model_callback,+ attacks=attacks,vulnerabilities=vulnerabilities)```@@ -331,8 +331,8 @@ vulnerabilities = owasp_asi.vulnerabilities# Modify attributes for your specific testing context if neededred_team(- model_callback=your_model_callback,- attacks=attacks,+ model_callback=your_model_callback,+ attacks=attacks,vulnerabilities=vulnerabilities)```@@ -363,8 +363,8 @@ vulnerabilities = owasp_asi.vulnerabilities# Modify attributes for your specific testing context if neededred_team(- model_callback=your_model_callback,- attacks=attacks,+ model_callback=your_model_callback,+ attacks=attacks,vulnerabilities=vulnerabilities)```@@ -446,6 +446,7 @@ The OWASP ASI Top 10 builds upon and extends the OWASP Top 10 for LLMs. Each age:::tip**Testing Strategy**: If you're building agentic systems, test against **both** frameworks:+- Use the **OWASP Top 10 for LLMs** to validate foundational model security- Use the **OWASP ASI Top 10** to assess agentic-specific risks like tool orchestration, inter-agent communication, and cascading failures@@ -481,4 +482,4 @@ DeepTeam tests agentic security risks at the **model reasoning and response leveFor **comprehensive agentic security**, you must implement additional runtime monitoring, access controls, and infrastructure security measures. Read the official [OWASP Top 10 for Agentic Applications 2026](https://genai.owasp.org/resource/owasp-top-10-for-agentic-applications-for-2026/) documentation to understand the complete security requirements for production agentic systems.**DeepTeam helps you test behavioral vulnerabilities at the model layer. Production agentic security requires defense-in-depth across all system layers.**-:::\ No newline at end of file+:::diff --git a/docs/docs/frameworks-owasp-top-10-for-llms.mdx b/docs/docs/frameworks-owasp-top-10-for-llms.mdxindex 712caf2b..75a04e08 100644--- a/docs/docs/frameworks-owasp-top-10-for-llms.mdx+++ b/docs/docs/frameworks-owasp-top-10-for-llms.mdx@@ -1,7 +1,7 @@---id: frameworks-owasp-top-10-for-llms-title: OWASP Top 10 for LLMs-sidebar_label: OWASP Top 10 for LLMs+title: OWASP Top 10 for LLMs 2025+sidebar_label: OWASP Top 10 for LLMs 2025---The **OWASP Top 10 for Large Language Models (LLMs)** is a comprehensive list of the most critical security risks associated with LLM applications. This resource is designed to help developers, security professionals, and organizations identify, understand, and mitigate vulnerabilities in these systems, ensuring safer and more robust deployments in real-world applications.@@ -73,8 +73,8 @@ vulnerabilities = owasp.vulnerabilities# Modify attributes for your specific testing context if neededred_team(- model_callback=your_model_callback,- attacks=attacks,+ model_callback=your_model_callback,+ attacks=attacks,vulnerabilities=vulnerabilities)```@@ -105,8 +105,8 @@ vulnerabilities = owasp.vulnerabilities# Modify attributes for your specific testing context if neededred_team(- model_callback=your_model_callback,- attacks=attacks,+ model_callback=your_model_callback,+ attacks=attacks,vulnerabilities=vulnerabilities)```@@ -137,8 +137,8 @@ vulnerabilities = owasp.vulnerabilities# Modify attributes for your specific testing context if neededred_team(- model_callback=your_model_callback,- attacks=attacks,+ model_callback=your_model_callback,+ attacks=attacks,vulnerabilities=vulnerabilities)```@@ -169,8 +169,8 @@ vulnerabilities = owasp.vulnerabilities# Modify attributes for your specific testing context if neededred_team(- model_callback=your_model_callback,- attacks=attacks,+ model_callback=your_model_callback,+ attacks=attacks,vulnerabilities=vulnerabilities)```@@ -197,8 +197,8 @@ vulnerabilities = owasp.vulnerabilities# Modify attributes for your specific testing context if neededred_team(- model_callback=your_model_callback,- attacks=attacks,+ model_callback=your_model_callback,+ attacks=attacks,vulnerabilities=vulnerabilities)```@@ -228,8 +228,8 @@ vulnerabilities = owasp.vulnerabilities# Modify attributes for your specific testing context if neededred_team(- model_callback=your_model_callback,- attacks=attacks,+ model_callback=your_model_callback,+ attacks=attacks,vulnerabilities=vulnerabilities)```@@ -256,8 +256,8 @@ vulnerabilities = owasp.vulnerabilities# Modify attributes for your specific testing context if neededred_team(- model_callback=your_model_callback,- attacks=attacks,+ model_callback=your_model_callback,+ attacks=attacks,vulnerabilities=vulnerabilities)```@@ -288,8 +288,8 @@ vulnerabilities = owasp.vulnerabilities# Modify attributes for your specific testing context if neededred_team(- model_callback=your_model_callback,- attacks=attacks,+ model_callback=your_model_callback,+ attacks=attacks,vulnerabilities=vulnerabilities)```@@ -320,8 +320,8 @@ vulnerabilities = owasp.vulnerabilities# Modify attributes for your specific testing context if neededred_team(- model_callback=your_model_callback,- attacks=attacks,+ model_callback=your_model_callback,+ attacks=attacks,vulnerabilities=vulnerabilities)```@@ -352,8 +352,8 @@ vulnerabilities = owasp.vulnerabilities# Modify attributes for your specific testing context if neededred_team(- model_callback=your_model_callback,- attacks=attacks,+ model_callback=your_model_callback,+ attacks=attacks,vulnerabilities=vulnerabilities)```diff --git a/docs/sidebars.js b/docs/sidebars.jsindex 8b81e708..67ddf52f 100644--- a/docs/sidebars.js+++ b/docs/sidebars.js@@ -166,8 +166,15 @@ module.exports = {className: "sidebar-item-guidelines-frameworks root-sidebar-item",items: ["frameworks-introduction",- "frameworks-owasp-top-10-for-llms",- "frameworks-owasp-top-10-for-agentic-applications",+ {+ type: "category",+ label: "OWASP",+ items: [+ "frameworks-owasp-top-10-for-llms",+ "frameworks-owasp-top-10-for-agentic-applications",+ ],+ collapsed: false,+ },"frameworks-nist-ai-rmf","frameworks-mitre-atlas","frameworks-beavertails",