You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.
Use when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies
Use when you have a written implementation plan to execute in a separate session with review checkpoints
Use when implementation is complete, all tests pass, and you need to decide how to integrate the work - guides completion of development work by presenting structured options for merge, PR, or cleanup
Use when receiving code review feedback, before implementing suggestions, especially if feedback seems unclear or technically questionable - requires technical rigor and verification, not performative agreement or blind implementation
Uses power tools
Uses Bash, Write, or Edit tools
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
A fork of obra/superpowers.
For the core concepts, workflow, skills library, and design philosophy, see the upstream README.
Adds a configurable mechanism to dispatch code reviews to different AI providers (Codex CLI, Claude Code, etc.). The provider is selected via review_provider in either ${XDG_CONFIG_HOME:-~/.config}/superpowers/review-config.json (user global; recommended for personal defaults) or .superpowers/review-config.json (project; overrides global per-project). Both files are optional; when both exist, project keys override global by key — with coding.rules as the documented exception (merged by category; see below). Provider definitions live in skills/requesting-code-review/providers/*.json. Falls back to host-AI subagents when the configured provider is unavailable.
git merge-base for stable diff boundariessuperpowers-multi:requesting-code-reviewreview-prompt.md (code quality) and spec-review-prompt.md (spec compliance) replace the previous Codex/Claude-specific pairRoutes implementation tasks to AI providers by task category, so frontend and backend work can be handled by different providers. Configured via the coding key in the same review-config.json files described above (global at ${XDG_CONFIG_HOME:-~/.config}/superpowers/, project at <repo>/.superpowers/). The coding.rules array merges by category: a project override of backend keeps the global frontend entry intact. An empty rules: [] in the project config explicitly disables global rules for that project.
{
"review_provider": "codex",
"coding": {
"enabled": true,
"default_provider": "codex",
"rules": [
{ "category": "frontend", "provider": "claude-code" },
{ "category": "backend", "provider": "codex" }
]
}
}
frontend / backend / fullstack) via plan tag or AI auto-classification, then routed to the configured provider. Results are validated (file changes, non-empty output, no timeout) before passing into the existing two-stage review. On failure or when disabled, falls back to the existing host implementercategory: field that overrides auto-classificationcoding-prompt.md is used by all providers; provider JSON files may add invoke_coding / plugin_override_coding for coding-specific CLI argscoding block is absent or coding.enabled is false, the existing implementer flow is unchanged. (coding.enabled defaults to true when omitted from a coding block, so configs that only set default_provider or rules remain active.)Works out of the box with no additional dependencies. For enhanced multi-AI dispatch:
${XDG_CONFIG_HOME:-~/.config}/superpowers/review-config.json (recommended for personal defaults across all projects) and/or project config at .superpowers/review-config.json (overrides global per-project). Both are optional. Without either, the system prompts to set one up the first time it runs and asks where to save (global / project / session-only).Register the marketplace, then install the plugin:
/plugin marketplace add undrthemt/superpowers-multi
/plugin install superpowers-multi@superpowers-multi
To update:
/plugin marketplace update superpowers-multi
/plugin install superpowers-multi@superpowers-multi
/plugin install reads from the local marketplace clone, not from GitHub directly, so the marketplace clone must be updated first. /reload-plugins only reloads the currently-installed version — it does not fetch new releases.
npx claudepluginhub undrthemt/superpowers-multiCore skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques
Superpowers Plus core skills library for Claude Code: planning, execution routing, TDD, debugging, and collaboration workflows
v9.52.0 - Reliability wave: tangle contextual review correction loop with hard round ceiling, progress-supervised review rounds (per-agent stall watch, descendant-tree kills), council diversity and agy pin fixes, marketplace generator source-of-truth fix, provider troubleshooting runbook and cost-expectations docs. Run /octo:setup.
Evidence-gated AI coding workflow: scan → analyze → plan → TDD → execute → fix → verify → review, powered by Codebase Memory MCP >= 0.9.0 with optional Serena LSP intelligence. Includes blast-radius planning, test/cycle gates, independent review, and Windows Git Bash hook auto-resolution.
Claude harness - A harness for solo developers (Vibecoders) to handle full-cycle contract development.
Persistent file-based planning for AI coding agents. Crash-proof markdown plans (task_plan.md, findings.md, progress.md) that survive context loss and /clear, with an opt-in completion gate and multi-agent shared state. Manus-style. Works with Claude Code, Codex CLI, Cursor, Kiro, OpenCode and 60+ agents via the SKILL.md standard. Includes Arabic, German, Spanish, and Chinese (Simplified and Traditional).