Core skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques
Use when designing new functionality, creating features, building components, adding user-facing behavior, or changing product requirements 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, tests pass, or work is ready for merge, PR, cleanup, handoff, or branch integration
Use when receiving, evaluating, or responding to code review feedback, reviewer comments, requested changes, or technical criticism
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.
Codex Superpowers is a Codex-first fork of obra/superpowers.
The upstream project provides the software-development discipline: skill
selection, brainstorming, planning, TDD, systematic debugging, subagent-driven
development, code review, and branch finishing. This fork keeps that discipline
on the upstream v6.1.1 skill base, then adds a Codex overlay so the active
instructions match Codex's local skill discovery and tools.
This repository is not the original Superpowers project. It is a public fork optimized for Codex users.
v6.1.1For other harnesses, use the upstream repository and its installation docs.
Clone this fork:
git clone https://github.com/smallocean43658/codex-superpowers.git ~/.codex/superpowers
Expose the skills to Codex:
mkdir -p ~/.agents/skills
ln -s ~/.codex/superpowers/skills ~/.agents/skills/superpowers
Restart Codex so it discovers the skills.
This fork is intended to replace the active superpowers skills entry in
Codex, not run beside the upstream marketplace install. If
~/.agents/skills/superpowers already exists, inspect it and replace it with
the symlink above:
ls -la ~/.agents/skills/superpowers
rm ~/.agents/skills/superpowers
ln -s ~/.codex/superpowers/skills ~/.agents/skills/superpowers
Use the upstream marketplace package if you want the official multi-harness Superpowers plugin. Use this fork when you want the Codex-specific local skill overlay.
ls -la ~/.agents/skills/superpowers
You should see a symlink pointing to:
~/.codex/superpowers/skills
Then start a new Codex session and ask for something that should trigger a skill:
help me plan this feature
or:
let's debug this failing test
Codex should load and announce the relevant Superpowers skill.
Subagent-heavy workflows such as dispatching-parallel-agents and
subagent-driven-development require Codex multi-agent support. Add this to
~/.codex/config.toml:
[features]
multi_agent = true
Without this feature, the planning, debugging, TDD, and review disciplines still apply, but subagent dispatch instructions need to be executed inline.
cd ~/.codex/superpowers
git pull
Because Codex reads skills through the symlink, updates take effect after a new Codex session starts.
This fork keeps upstream v6.1.1 skill behavior, including the newer
subagent-driven development file handoffs, durable progress ledger, worktree
detection, and hardened brainstorming visual companion.
The Codex overlay adds:
update_plan, spawn_agent, wait_agent,
close_agent, apply_patch, and exec_commandusing-superpowers entry guidanceSee CODEX_OPTIMIZATIONS.md for the detailed notes.
using-superpowers - entry discipline document for checking and loading skillsbrainstorming - design new functionality before implementationusing-git-worktrees - ensure isolated workspace behavior is deliberatewriting-plans - turn approved specs into implementation planssubagent-driven-development - execute independent plan tasks with task
review and final whole-branch reviewexecuting-plans - execute written plans inline or in a separate sessiontest-driven-development - enforce RED/GREEN/REFACTORsystematic-debugging - find root cause before fixingrequesting-code-review - request independent review before proceedingreceiving-code-review - evaluate review feedback rigorouslyverification-before-completion - verify before claiming work is completefinishing-a-development-branch - finish, verify, and decide merge/PR/handoffThe Codex overlay is checked with:
tests/codex/test-codex-fork-overlay.shtests/codex/test-marketplace-manifest.shgit diff --checkThis repository keeps the upstream MIT license. See LICENSE.
npx claudepluginhub smallocean43658/codex-superpowers --plugin superpowersCore 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
Harness-native ECC operator layer - 67 agents, 277 skills, 92 legacy command shims, reusable hooks, rules, selective install profiles, and production-ready workflows for Claude Code, Codex, OpenCode, Cursor, and related agent harnesses
Supergraph enforces a complete, evidence-based coding pipeline — scan → plan → TDD → fix → verify → review — grounded in real codebase analysis at every step. It combines AST dependency graphs, LSP-level code intelligence, and a structured skill chain so Claude never guesses about impact before making a change.
Claude harness - A harness for solo developers (Vibecoders) to handle full-cycle contract development.
Plugin-safe Claude Code distribution of Agentic Awesome Skills with 1,894 supported skills.