What problem does it solve? Documents written for AI agents often fail because agents under-trigger on weak pointers, rush steps with vague completion criteria, or drown in bloated files. This Skill provides a systematic reference for writing any document an agent consumes so it behaves predictably on every run. ## Core Features & Use Cases - Context pointer design: Teaches how to word skill descriptions and AGENTS.md lines so agents reliably reach the right material at the right branch. - Information hierarchy and progressive disclosure: Explains the ladder of in-file steps, in-file reference, and disclosed reference, plus co-location, sprawl, and when to split documents by sequence or invocation. - Completion criteria and pruning: Covers writing checkable, demanding step boundaries, using leading words, avoiding negation, and removing duplication, caches, no-ops, and sediment. - Use Case: When creating a new skill or editing an AGENTS.md file, apply these rules to sharpen the description's trigger branches, push branch-specific reference into a references/ file, and give every step a clear done condition. ## Quick Start Ask the agent to review your draft skill or AGENTS.md file using the writing-for-agents guidance and rewrite weak pointers and vague completion criteria.
