Claude Code plugin for interview-driven implementation plans — /plan-md interrogates you until every design decision is closed, then writes a plan an AI agent can execute cold.
This repository also configures its own agents. See what claude-plugins tells them →
Plugin Claude Code
Bundles 6 skills, 1 command, 2 hooks · 262 tokens together
Project intelligence layer manager — bootstrap, extend, and audit an intelligence/index.md index + intelligence/.md topic files.
Plugin Claude Code
Bundles 7 skills, 1 command · 292 tokens together
Interview-driven implementation plans written for AI coding agents — create, list, review, execute.
Plugin Claude Code
Bundles 3 skills, 2 commands · 161 tokens together
AI pull-request reviews — review a GitHub PR (or your own current branch via /self-review), then plan the fixes, keep a local report, or push inline review comments (beta).
Plugin Claude Code
Bundles 1 command · 19 tokens together
Merge a branch into the current branch, resolving conflicts intelligently — defaults to the remote default branch.
Plugin Claude Code
Bundles 1 skill · 56 tokens together
Deny code comments at write time — Claude rewrites the code with descriptive names and small named functions so the comment is redundant.
Plugin Claude Code
Bundles 1 command · 37 tokens together
Prime a repository for quality software engineering — installs direction-focused intelligence topics for Arrange–Act–Assert tests, Clean Code, Clean Architecture, comment-free code, the boy scout rule, and the detected stack.
Plugin Claude Code
Bundles 1 skill, 1 command · 125 tokens together
Writing rules for all output — chat replies, review comments, docs, changelogs, commit and PR bodies: lead with the point, state only what is in place, plain language per ISO 24495-1.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: