What problem does it solve? Writing agent skills that behave consistently is hard: descriptions misfire, instructions bloat, and agents rush steps or ignore guidance. This Skill supplies a shared vocabulary and set of principles for designing, structuring, and pruning skills so they behave predictably on every run. ## Core Features & Use Cases - Invocation Design: Explains the trade-off between model-invoked skills (agent-discoverable, with context load) and user-invoked skills (zero context load, human-triggered), including router skills for managing many user-invoked ones. - Information Hierarchy: Defines how to arrange steps and reference material across SKILL.md and disclosed files using progressive disclosure and context pointers. - Failure Mode Diagnosis: Catalogs failure modes such as premature completion, duplication, sediment, sprawl, no-ops, and negation, each paired with its cure. - Use Case: When editing an existing skill that the agent triggers unreliably, consult this reference to sharpen the description's trigger wording, collapse duplicated branches, and push rarely needed reference material behind a pointer. ## Quick Start Ask the agent to review your SKILL.md using the writing-great-skills principles and suggest improvements to its description and structure.
