Model behavior and evaluation
Explore model capabilities, reasoning, factuality, sycophancy, and deception, with research and methods for evaluating AI answers.
AI is powerful when used intelligently.
Explore the research behind AI models and agent systems—from behavior to security. Put that understanding to work with practical guides and a library of ready-to-use prompts.
Explore the knowledge hub
Choose a topic to find research articles, practical guides, prompts, and reference material.
-
-
AI security
Investigate prompt injection, jailbreaks, trust boundaries, permissions, and data exposure across AI models, agents, and connected tools.
-
Architecture and integrations
Understand and design agents, RAG, memory, orchestration, APIs, MCP, and tool integrations, including their control and security boundaries.
-
Prompt engineering
Write and assemble prompts with clear instructions, source boundaries, reusable components, output requirements, and verification steps.
-
Using AI tools
Use ChatGPT, Claude, and Gemini for research, writing, and everyday work. Find guidance on files, memory, browsing, configuration, and checking results.
-
AI-assisted software development
Generate and review code with AI, check architecture and implementation, plan refactors, and evaluate tests and regression risks.
Featured technical articles
Research and technical analysis of AI behavior, deception, jailbreaks, and coding risks.
The company behind the knowledge hub
Andy Agent Lab is an AI research and product development company founded and operated by Tamar Peretz. The company studies how AI reproduces patterns of human cognition, emotion, and social behavior, connecting this research with product development and offensive and defensive AI security.
Latest updates
Recent additions to guides, prompts, policies, articles, reference pages, and site navigation.