Progressive Disclosure
Progressive disclosure is an information-presentation strategy in which complex content is layered and revealed on demand. Basic or high-frequency information is shown first, with additional details exposed only when a user explicitly requests them or specific conditions are met. The principle is widely applied in user interface design, instructional design, and system interaction to reduce cognitive load and information overload while preserving the completeness and accessibility of the full content.
Core Facts
Progressive Disclosure is originally a classic principle in user interface design, referring to showing complex information only when the user needs it.
The Skill mechanism uses on-demand loading: it first sends the metadata of all Skills to the large model, and after the large model selects the matching Skill, the full instructions are loaded.
Skills采用渐进式披露设计，平时只向模型提供技能的名称和描述，具体执行规则在需要时才加载
Timeline (last 90 days)
Progressive disclosure advocates loading context only when truly needed; CLAUDE.md should remain lightweight and serve as an index, replacing large inline content with file paths.
Anthropic's progressive disclosure mechanism works as follows: at startup, only the name and description of each skill (YAML front matter) are read; when the prompt matches a skill, the full skill MD file is read; larger reference files and scripts are loaded on demand.
Excellent interface design should follow the principle of Progressive Disclosure, presenting information to users only when they need it
Progressive Disclosure is a compromise approach: only the most likely root cause is shown by default, but an option to expand the full chain is provided, and Kubernetes's kubectl command adopts a similar design.
Progressive Disclosure is originally a classic principle in user interface design, referring to showing complex information only when the user needs it.
All Facts (8)
Progressive Disclosure is originally a classic principle in user interface design, referring to showing complex information only when the user needs it.
75%VerifiedThe Skill mechanism uses on-demand loading: it first sends the metadata of all Skills to the large model, and after the large model selects the matching Skill, the full instructions are loaded.
65%VerifiedSkills采用渐进式披露设计，平时只向模型提供技能的名称和描述，具体执行规则在需要时才加载
65%UnverifiedSkill由三层结构组成：概要层（约100字）、主体层（约3000字）和细节层，采用渐进式披露策略
85%UnverifiedProgressive disclosure advocates loading context only when truly needed; CLAUDE.md should remain lightweight and serve as an index, replacing large inline content with file paths.
50%UnverifiedAnthropic's progressive disclosure mechanism works as follows: at startup, only the name and description of each skill (YAML front matter) are read; when the prompt matches a skill, the full skill MD file is read; larger reference files and scripts are loaded on demand.
50%UnverifiedExcellent interface design should follow the principle of Progressive Disclosure, presenting information to users only when they need it
50%UnverifiedProgressive Disclosure is a compromise approach: only the most likely root cause is shown by default, but an option to expand the full chain is provided, and Kubernetes's kubectl command adopts a similar design.
50%