# AI Agents and Patterns Hub
> [!Note]
> Sub-hub for the model-side and pattern-side of AI agents — what an
> agent is, the prompting techniques that drive it, the loops that make
> it autonomous, and the retrieval systems that ground it.
## 1. The agent itself
- [[AI Agent]]
- [[Agentic Loop]]
- [[Multi-Agent System]]
- [[Orchestrator-Subagent Pattern]]
- [[Subagent Context Isolation]]
- [[Compressed Summary Return]]
- [[Ephemeral Subagent]]
---
## 2. Driving the model
- [[Prompt Engineering]]
- [[System Prompt]]
- [[In-Context Learning]]
- [[Chain-of-Thought]]
---
## 3. Action and reasoning patterns
- [[Tool Use]]
- [[ReAct Pattern]]
- [[Reflection]]
---
## 4. Retrieval and grounding
- [[Retrieval-Augmented Generation]]
- [[Vector Database]]
- [[Semantic Search]]
- [[Embedding-Based Retrieval]]
---
## 5. Threats and failure modes
- [[Prompt Injection]]
- [[Hallucination]]
---
## 6. Anchor sources
- [[Chain-of-Thought Prompting (Wei et al.)]]
- [[ReAct (Yao et al.)]]
- [[Retrieval-Augmented Generation (Lewis et al.)]]
- [[Building Effective AI Agents]]
- [[Multi-Agent AI Systems in 2026 (FlowHunt)]]
- [[The Architecture of Scale - Anthropic Sub-Agents (Oswal)]]
---
## Up
- [[0 - Modern AI Software Engineering Hub]]