Overview
Feedback given in chat gets lost. "Revise the part about the database" means nothing a week later, in a new session or on another device. With this skill, you write feedback directly in the document as %% markers %%. Claude edits the content and replies next to your comment. Everything stays in plain markdown, so git keeps the whole history.
What it does
- Inline comments:
%% comment %%asks for a change,%% ?: question %%asks a question,%% NOTE: context %%gives context, and%% WIP %%marks unfinished sections - Precise targets: mark text with
==PostgreSQL(DB)==, then refer to it by name:%%(DB) switch to SQLite %% - Inline replies: Claude answers with
•%%> done <%%•, so the comment, the reply and the updated text stay together - A simple workflow: explore in chat, draft the document, iterate with markers, then remove the markers and keep the content
- Works beyond Claude Code: it's built for Claude but works with any AI that follows structured instructions. Use it as a Claude Code skill, or add
SKILL.mdto a Claude.ai Project.
The full syntax, examples and FAQ are on the docs site.
Built with
Markdown and a SKILL.md skill definition. There's nothing to install.