superpowers-plus

Superpowers Plus

English | 中文

Superpowers Plus is an enhanced workflow skill set evolved from upstream superpowers, designed for coding agent environments like Claude Code, Codex, OpenCode, and Gemini CLI. It retains the "design → plan → execute → verify" core workflow but unifies execution routing, subagent context control, and installation entry points within this repository instead of relying on the upstream.

How it works

It starts from the moment you fire up your coding agent. As soon as it sees that you're building something, it doesn't just jump into trying to write code. Instead, it steps back and asks you what you're really trying to do.

Once it's teased a spec out of the conversation, it shows it to you in chunks short enough to actually read and digest.

After you've signed off on the design, your agent puts together an implementation plan that's clear enough for an enthusiastic junior engineer with poor taste, no judgement, no project context, and an aversion to testing to follow. It emphasizes true red/green TDD, YAGNI (You Aren't Gonna Need It), and DRY.

Next up, once you say "go", it launches an executing-plans process that automatically chooses parallel or serial subagent routing based on task dependency, then inspects and reviews the work as it goes. It's not uncommon for Claude to be able to work autonomously for a couple hours at a time without deviating from the plan you put together.

There's a bunch more to it, but that's the core of the system. And because the skills trigger automatically, you don't need to do anything special. Your coding agent just has Superpowers Plus.

Key Changes from Upstream Superpowers

Superpowers Plus has made the following consolidations and adjustments relative to the upstream:

Execution Model Unification:

• Unified the plan execution entry point to executing-plans, no longer maintaining subagent-driven-development as a separate workflow entry.
• executing-plans now operates as a controller-style executor: automatically determines whether tasks are suitable for parallel subagents or require serial subagents, eliminating the need for users to choose between the two before execution.
• Subagents now receive only task-level minimal context by default, not the entire conversation history; execution models default to lower-cost sub-models, upgrading only when necessary.
• Plan handoff messaging now leads directly to execution, no longer inserting a "Subagent-Driven / Inline Execution" choice after writing-plans.

Simplified Subagent Workflow:

• Removed mandatory spec and code review requirements for subagent execution.
• Eliminated upstream's prescriptive subagent development conventions.
• Design philosophy: Since brainstorming and writing-plans already provide thorough design and planning, the execution phase should trust and follow that plan rather than repeat design processes at the subagent level.
• Enforces frequent, granular commits: every RED-GREEN-REFACTOR cycle commits independently, creating a clear implementation audit trail.

Enhanced Documentation Management:

• Introduced structured documentation tracking system:
  • Design specs → docs/superpowers/specs/YYYY-MM-DD-<topic>-design.md
  • Implementation plans → docs/superpowers/plans/YYYY-MM-DD-<feature-name>.md
  • Commit documentation → docs/superpowers/commits/<module>/
• Built-in self-review mechanisms for both specs and plans before user review, catching placeholders, contradictions, and ambiguities early.
• Strengthened repository-level documentation indexing in AGENTS.md and CLAUDE.md.

Repository Consistency:

• Tests, examples, and active documentation within the repository have been updated around the new single-entry execution model, avoiding discrepancies between README, skills, and tests.

This is not a complete changelog, but rather the behavioral differences users need to know.

Installation

Superpowers Plus installation entry points are unified to this repository:

• GitHub: https://github.com/xhyqaq/superpowers-plus
• Codex install script: https://raw.githubusercontent.com/xhyqaq/superpowers-plus/refs/heads/main/.codex/INSTALL.md
• OpenCode install script: https://raw.githubusercontent.com/xhyqaq/superpowers-plus/refs/heads/main/.opencode/INSTALL.md

Claude Code

We recommend installing directly from the repository rather than relying on the upstream marketplace:

Quick Install (Recommended):

git clone https://github.com/xhyqaq/superpowers-plus.git ~/.claude/superpowers-plus
bash ~/.claude/superpowers-plus/.claude-plugin/install.sh

Manual Install:

git clone https://github.com/xhyqaq/superpowers-plus.git ~/.claude/superpowers-plus
mkdir -p ~/.claude/skills
cd ~/.claude/superpowers-plus/skills
for skill in */; do
  skill_name="${skill%/}"
  ln -sf "$HOME/.claude/superpowers-plus/skills/$skill_name" "$HOME/.claude/skills/superpowers:$skill_name"
done