Skip to content
Last Dictionary
Plane
Agents & HarnessesAI Engineering & MLOps

progressive disclosure

noun

In plain wordsShowing only a short summary first and revealing details when they are actually needed. AI agents use it to avoid filling their limited memory with instructions they may never use.

  1. 1

    Agentsemerging

    A context engineeringcontext engineering, n.: Managing an AI model's reading material: choosing, trimming and arranging the instructions, documents, tool results and memories it is given on every turn, so that it has what the task needs and little else. pattern in which an AI agentAI agent, n.: An AI system that does not just answer but acts: given a goal, it decides on steps, uses tools such as a browser or code editor, checks the results and keeps going until the job is done. is given only brief descriptions of available resources — names and one-line summaries of skillsagent skill, n.: A packaged how-to guide for an AI agent — written steps, sometimes with small programs and examples attached — that sits unread until a job of that kind comes up, when the agent opens it., toolstool, n.: Any outside capability an AI model is allowed to use — a web search, a calculator, a code runner, a database or an app — described to the model so it can ask for it by name. or documents — and loads fuller layers only when a task calls for them. In Anthropic's Agent Skills design, for example, the first level is each skill's name and description in the system promptsystem prompt, n.: Background instructions given to a chatbot before the conversation starts, setting its role, rules, tone and tools. Users usually never see it, but it shapes every reply.; the second is the skill's full instruction file, read when the skill looks relevant; deeper levels are linked scripts and reference files opened as needed. The pattern gives an agent access to far more capability than its context windowcontext window, n.: How much an AI model can read and keep in view at once: the instructions, the conversation so far and any attached documents. Text that does not fit has to be cut, summarized or left out. could hold at once and keeps irrelevant instructions from distracting it, but it depends on summaries good enough for the model to know when to look deeper.

    Thanks to progressive disclosure, installing forty skills costs only a few hundred tokens until one is used.

  2. 2

    Interface design

    In interface design, the practice, much promoted by usability writer Jakob Nielsen, of presenting the few functions most people need at first and tucking rarer or more advanced ones behind menus, links or 'advanced' panels until a user goes looking, so that beginners face a simple screen while experts keep full power.

Origin

A long-established principle of human–computer interaction, set out in an often-cited 2006 article by Jakob Nielsen; who first named it is not established. Anthropic applied it to agents in its September 2025 context-engineering guide and called it the core design principle of Agent Skills (16 October 2025); many agent harnesses now load skills, tools and documentation this way.

Related

See also
  • context engineeringcontext engineering, n.: Managing an AI model's reading material: choosing, trimming and arranging the instructions, documents, tool results and memories it is given on every turn, so that it has what the task needs and little else.
  • agent skillagent skill, n.: A packaged how-to guide for an AI agent — written steps, sometimes with small programs and examples attached — that sits unread until a job of that kind comes up, when the agent opens it.
  • just-in-time retrievaljust-in-time retrieval, n.: Letting an AI agent fetch information only at the moment it needs it — opening a file, running a search — instead of stuffing everything into its prompt in advance.
  • system promptsystem prompt, n.: Background instructions given to a chatbot before the conversation starts, setting its role, rules, tone and tools. Users usually never see it, but it shapes every reply.
  • context windowcontext window, n.: How much an AI model can read and keep in view at once: the instructions, the conversation so far and any attached documents. Text that does not fit has to be cut, summarized or left out.
  • tool usetool use, n.: An AI model's ability to reach beyond its own knowledge by operating outside tools — searching the web, running code, using a calculator or an app — and folding the results into its work.

Sources

  1. Anthropic (2025). Equipping agents for the real world with Agent Skills. Anthropic Engineering blog, 16 October 2025. anthropic.com/engineering/equipping-agents-for-the-real-world-with-agent-skills
  2. Nielsen, J. (2006). Progressive Disclosure. Nielsen Norman Group, 3 December 2006. nngroup.com/articles/progressive-disclosure

Mentioned in

agent skillagent skill, n.: A packaged how-to guide for an AI agent — written steps, sometimes with small programs and examples attached — that sits unread until a job of that kind comes up, when the agent opens it., context engineeringcontext engineering, n.: Managing an AI model's reading material: choosing, trimming and arranging the instructions, documents, tool results and memories it is given on every turn, so that it has what the task needs and little else., code modecode mode, n.: A way for AI agents to use tools by writing a short program that calls them, instead of asking for each tool one step at a time., just-in-time retrievaljust-in-time retrieval, n.: Letting an AI agent fetch information only at the moment it needs it — opening a file, running a search — instead of stuffing everything into its prompt in advance., llms.txtllms.txt, n.: A short guide file that a website puts at a standard address to help AI assistants find and understand its most useful pages.