24  Appendix C — Further Reading

A curated starting point. Full details are in the References.

24.1 Foundations

• Russell & Norvig, Artificial Intelligence: A Modern Approach, 4th ed. [1]
• Wooldridge & Jennings, Intelligent Agents: Theory and Practice [2]
• Sutton & Barto, Reinforcement Learning: An Introduction [3]

24.2 Reasoning & planning

• Wei et al., Chain-of-Thought Prompting [4]
• Yao et al., ReAct [5]
• Yao et al., Tree of Thoughts [6]
• Shinn et al., Reflexion [7]

24.3 Tools, memory, multi-agent

• Schick et al., Toolformer [8]
• Lewis et al., Retrieval-Augmented Generation [9]
• Park et al., Generative Agents [10]
• Wang et al., Voyager [11]

24.4 Industry guidance

• Anthropic, Building Effective Agents [12]
• OpenAI, New Tools for Building Agents [13]
• Model Context Protocol [14]
• LangGraph documentation [15]

24.5 Evaluation & safety

• AgentBench [16]
• SWE-bench [17]
• SWE-Bench Pro — professional-scale coding benchmark [18]
• GAIA — General AI Assistants benchmark [19]
• WebArena — reproducible web-agent benchmark [20]
• OSWorld — real-computer agent benchmark [21]
• AgentDojo — dynamic prompt-injection evaluation [22]
• OWASP Top 10 for LLM Applications [23]

24.6 2025 frontier work

• METR, Measuring AI Ability to Complete Long Tasks — the “task horizon doubles every ~7 months” study, on the ceiling of long-horizon agent reliability [24]
• Debenedetti et al., Defeating Prompt Injections by Design (CaMeL) [25]
• EU AI Act, Regulation (EU) 2024/1689 — the first comprehensive AI regulation and its risk tiers [26]
• Anthropic, Responsible Scaling Policy — AI Safety Levels for frontier models [27]

[1]

S. Russell and P. Norvig, Artificial intelligence: A modern approach, 4th ed. Pearson, 2021.

[2]

M. Wooldridge and N. R. Jennings, “Intelligent agents: Theory and practice,” The Knowledge Engineering Review, vol. 10, no. 2, pp. 115–152, 1995.

[3]

R. S. Sutton and A. G. Barto, Reinforcement learning: An introduction, 2nd ed. MIT Press, 2018.

[4]

J. Wei et al., “Chain-of-thought prompting elicits reasoning in large language models,” in Advances in neural information processing systems (NeurIPS), 2022.

[5]

S. Yao et al., “ReAct: Synergizing reasoning and acting in language models,” in International conference on learning representations (ICLR), 2023.

[6]

S. Yao et al., “Tree of thoughts: Deliberate problem solving with large language models,” in Advances in neural information processing systems (NeurIPS), 2023.

[7]

N. Shinn, F. Cassano, E. Berman, A. Gopinath, K. Narasimhan, and S. Yao, “Reflexion: Language agents with verbal reinforcement learning,” in Advances in neural information processing systems (NeurIPS), 2023.

[8]

T. Schick et al., “Toolformer: Language models can teach themselves to use tools,” in Advances in neural information processing systems (NeurIPS), 2023.

[9]

P. Lewis et al., “Retrieval-augmented generation for knowledge-intensive NLP tasks,” in Advances in neural information processing systems (NeurIPS), 2020.

[10]

J. S. Park, J. C. O’Brien, C. J. Cai, M. R. Morris, P. Liang, and M. S. Bernstein, “Generative agents: Interactive simulacra of human behavior,” in ACM symposium on user interface software and technology (UIST), 2023.

[11]

G. Wang et al., “Voyager: An open-ended embodied agent with large language models,” arXiv preprint arXiv:2305.16291, 2023.

[12]

Anthropic, E. Schluntz, and B. Zhang, “Building effective agents.” https://www.anthropic.com/engineering/building-effective-agents, Dec. 2024.

[13]

OpenAI, “New tools for building agents.” https://openai.com/index/new-tools-for-building-agents/, Mar. 2025.

[14]

Anthropic, “Model context protocol.” https://modelcontextprotocol.io, 2024.

[15]

LangChain, “LangGraph documentation.” https://langchain-ai.github.io/langgraph/, 2024.

[16]

X. Liu, H. Yu, H. Zhang, et al., “AgentBench: Evaluating LLMs as agents,” in International conference on learning representations (ICLR), 2024.

[17]

C. E. Jimenez et al., “SWE-bench: Can language models resolve real-world GitHub issues?” International Conference on Learning Representations (ICLR), 2024.

[18]

X. Deng et al., “SWE-Bench Pro: Can AI agents solve long-horizon software engineering tasks?” 2025. Available: https://arxiv.org/abs/2509.16941

[19]

G. Mialon, C. Fourrier, T. Wolf, Y. LeCun, and T. Scialom, “GAIA: A benchmark for general AI assistants,” in International conference on learning representations (ICLR), 2024. Available: https://arxiv.org/abs/2311.12983

[20]

S. Zhou et al., “WebArena: A realistic web environment for building autonomous agents,” in International conference on learning representations (ICLR), 2024. Available: https://arxiv.org/abs/2307.13854

[21]

T. Xie et al., “OSWorld: Benchmarking multimodal agents for open-ended tasks in real computer environments,” in Advances in neural information processing systems (NeurIPS), datasets and benchmarks track, 2024.

[22]

E. Debenedetti, J. Zhang, M. Balunović, L. Beurer-Kellner, M. Fischer, and F. Tramr, “AgentDojo: A dynamic environment to evaluate prompt injection attacks and defenses for LLM agents,” in Advances in neural information processing systems (NeurIPS), 2024. Available: https://arxiv.org/abs/2406.13352

[23]

OWASP, “OWASP top 10 for large language model applications.” https://owasp.org/www-project-top-10-for-large-language-model-applications/, 2025.

[24]

T. Kwa, B. West, J. Becker, et al., “Measuring AI ability to complete long tasks,” arXiv preprint arXiv:2503.14499, Mar. 2025, Available: https://metr.org/blog/2025-03-19-measuring-ai-ability-to-complete-long-tasks/

[25]

E. Debenedetti et al., “Defeating prompt injections by design,” arXiv preprint arXiv:2503.18813, 2025, Available: https://arxiv.org/abs/2503.18813

[26]

European Union, “Regulation (EU) 2024/1689 of the european parliament and of the council laying down harmonised rules on artificial intelligence (AI act).” https://eur-lex.europa.eu/eli/reg/2024/1689/oj, 2024.

[27]

Anthropic, “Anthropic’s responsible scaling policy.” https://www.anthropic.com/rsp, 2024.