Language: 日本語English

explanatory-output-style Plugin

Author
Anthropic
Category
Learning
Topics
Claude Code Customization & Workflow
First seen
2026-07-09
Last confirmed
2026-07-13
Explanation last updated
2026-07-12

The explanation below is AI-generated. Please verify it against the sources.

This plugin recreates the deprecated "Explanatory" output style using a SessionStart hook that injects instructions at the start of every session. It encourages Claude to provide educational insights about implementation choices, explain codebase patterns and decisions, and balance task completion with learning opportunities. These insights appear as brief formatted blocks (marked with a '★ Insight' header) before and after writing code, containing 2-3 key educational points. The README notes this incurs additional token cost from the extra instructions and output, and warns not to install it unless that trade-off is acceptable.

Overview

According to the README, this is a Claude Code plugin (not a SaaS) that adds a SessionStart hook, replacing the deprecated "Explanatory" output style setting for users who previously configured {"outputStyle": "Explanatory"}.

What you can do with explanatory-output-style

  • Automatically inject educational context at the start of every Claude Code session (no extra configuration needed)
  • Receive brief '★ Insight' blocks with 2-3 educational points before/after code changes
  • Get explanations focused on specific implementation choices, patterns/conventions, and trade-offs in your own codebase (per the README, rather than general programming concepts)
  • Migrate from the deprecated outputStyle: "Explanatory" setting to this plugin
  • Disable, uninstall, or create a personalized local copy of the plugin to manage its behavior

Sources

Original description (English)

Adds educational insights about implementation choices and codebase patterns (mimics the deprecated Explanatory output style)

History of explanatory-output-style

Back to list