planning
Having an agent write out the steps of a task before acting, and revise the plan as results come in.
In more detail
A written plan keeps a long job on course and can be saved outside the context. Check each step's result rather than trusting the plan, and let the agent replan when a step fails.
Related terms
- AI agent — A system in which a model decides its own next steps, calling tools in a loop until a task is done, rather than following a path your code fixed.
- agent memory — Information an AI application keeps outside the context window, such as user preferences or task notes, and loads back when relevant.
Taught in
- Workflows Before Agents — AI Engineering: LLM Apps, RAG & Agents
- Planning, Memory And Many Agents — AI Engineering: LLM Apps, RAG & Agents
- Threat-Model An LLM App — AI Engineering: LLM Apps, RAG & Agents
- Prompt Injection — AI Engineering: LLM Apps, RAG & Agents
References
- Anthropic: How we built our multi-agent research system
- Yao et al. (2022): ReAct: Synergizing Reasoning and Acting in Language Models