doc-organization

Reviews technical documentation for information placement, duplication, and missing context issues.

Updated Jan 12, 2026
One-click install
npx skills add https://github.com/ab300819/keel-workflow --skill doc-organization-ab300819
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: doc-organization
Source: https://github.com/ab300819/keel-workflow/tree/main/skills/doc-organization
Command: npx skills add https://github.com/ab300819/keel-workflow --skill doc-organization-ab300819

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Technical documents often mix change history with current state, scatter the same conclusion across multiple files, use undefined identifiers, and assume reader context that no longer exists. This Skill reviews Markdown documentation against five information-organization principles and explains where content belongs, without rewriting prose. ## Core Features & Use Cases - Information Placement Review: Judges whether change history, decision rationale, and current-state statements are correctly separated, and whether introductions serve as reading guides rather than content dumps. - Duplication and Definition Checks: Identifies conclusions defined in multiple places, numbered items used before being defined, and jargon that lacks context for future readers. - Use Case: After drafting a system design document, ask for a review to find sections where a deprecated plan is referenced by number but never defined, or where the same architectural conclusion is copied into three files and will drift. ## Quick Start Review this Markdown design document and point out any information organization problems such as duplicated conclusions, undefined identifiers, or change history mixed into the current-state description.

Frequently Asked Questions about doc-organization

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I review technical documentation for structural problems?

Review documentation against information organization principles: keep current state separate from change history, define identifiers before use, define each conclusion in exactly one place, and write for readers who lack your current context. This Skill applies those five principles as judgment, not mechanical rules.

What is the difference between document organization review and Markdown linting?

Markdown linting checks markup usage such as heading levels, code block languages, and punctuation spacing. Document organization review checks where information lives: whether conclusions are duplicated across files, whether introductions guide reading decisions, and whether deprecated items are left undefined.

Should duplicated content in documentation always be deduplicated?

No. Deliberate duplication can be justified, such as test cases quoting requirement text for local readability or runbooks staying self-contained offline. The requirement is knowing the synchronization cost you are paying, not mechanically removing every repetition.

Can AI fix undefined identifiers in existing documentation?

No. When a document references a numbered item that was never defined, no one can reconstruct what it pointed to after the fact, and AI would only be guessing. Definitions must be written at the moment the identifier is first introduced.

When should change history be kept out of a document's main body?

Change rationale belongs in a changelog or ADR section while the body states only the current position. Mixing them makes it impossible to tell which statement is authoritative, but rationale should stay near the content it explains rather than being dumped at the end.