Alanamultiskillagent
Back to skills

document-release

docs

Runs after code ships to bring all project documentation in sync with what actually changed. Reads the diff, then updates README, ARCHITECTURE, CONTRIBUTING, CLAUDE.md, and CHANGELOG to match the curr

garrytangarrytan
|View all garrytan skills →

Setup & Installation

npx skills add https://github.com/garrytan/document-release --skill document-release
or paste the link and ask your coding assistant to install it
https://github.com/garrytan/document-release
View on GitHub

What This Skill Does

Runs after code ships to bring all project documentation in sync with what actually changed. Reads the diff, then updates README, ARCHITECTURE, CONTRIBUTING, CLAUDE.md, and CHANGELOG to match the current state of the repo. Optionally bumps VERSION and cleans up completed TODOs. Instead of manually diffing each doc against the code after every merge, this traces the exact changes and applies only what the diff actually warrants, without overwriting content or guessing at intent.

When to use it

  • Working with document release functionality
  • Implementing document release features
  • Debugging document release related issues

Similar Skills

agents-md

A skill for creating and maintaining AGENTS.md files — the documentation that agent tools like Claude read when working in a project. It enforces a minimal format: under 60 lines, headers and bullets

content-modeling-best-practices

Structured guidance for designing content schemas in Sanity and other headless CMSes. Covers when to use references versus embedded objects, how to avoid page-shaped schemas, content reuse patterns, a

doc

Handles reading, creating, and editing .docx files where formatting fidelity matters. Uses python-docx for structured edits and a render pipeline (LibreOffice + Poppler) to visually inspect layout bef

doc-coauthoring

A structured workflow for co-authoring documents with Claude. It guides users through three stages: context gathering, section-by-section drafting with iterative refinement, and reader testing using a