Sources¶

Below is the core set of primary sources used in the current version of the book. Access date: March 29, 2026.

Agent architecture and patterns¶

• Dmitry Vikulin, “Architecture of Reliable AI Agents”
• Anthropic, Building Effective AI Agents
• LangGraph, Overview
• LangChain, Multi-agent
• OpenAI, Agents SDK
• OpenAI, Agent Builder

Reliability, memory, and HITL¶

• LangGraph, Durable execution
• LangGraph, Memory overview
• LangChain Deep Agents, Human-in-the-loop

Security and governance¶

• OWASP, LLM Prompt Injection Prevention Cheat Sheet
• NIST, AI RMF 1.0
• NIST, AI RMF: Generative AI Profile
• NIST, Adversarial Machine Learning: A Taxonomy and Terminology of Attacks and Mitigations
• Anthropic, Claude Code Security
• Google Cloud, Google Agentspace
• Google Cloud, Vertex AI Agent Builder

Observability and quality evaluation¶

• OpenAI, Agent evals
• OpenAI, Trace grading
• Google Cloud, Observability and monitoring

Publishing and tooling¶

• MkDocs, Official documentation
• Material for MkDocs, Official documentation
• uv, Working on projects
• ty, Official documentation
• Starlight, Official documentation

How to use this list¶

If you continue developing the book, the best reading order is:

1. Security and risk framing: NIST, OWASP.
2. Architectural patterns: Anthropic, LangGraph, OpenAI.
3. Governance and platform controls: Google Cloud, OpenAI, Anthropic.
4. Tooling and publishing: MkDocs, uv, ty, Starlight.