Eleven Claude Code workflows, adapted for modern OpenCode and Codex.
Code review, security audit, feature development, frontend design, project memory, and authoring tools.
Install · Skills · Invocation · Architecture · Credits
Why This Exists
OpenCode and Codex read SKILL.md workflows, but many valuable Claude Code workflows originated as Claude-specific commands and agents. Copying those artifacts directly does not preserve their orchestration, permissions, or subagent behavior.
This package adapts the portable methodology into shared skills, registers feature-development specialist roles as read-only OpenCode subagents, and lets Codex execute the same phase assignments with its native subagent workflow. It ships immutable provenance for every upstream work.
It complements obra/superpowers, which provides process skills such as brainstorming, TDD, debugging, and plan execution.
What's Inside
| Category | Skill | Purpose |
|---|---|---|
| Review | code-review |
Multi-agent PR review with confidence filtering and reproduction scenarios |
| Review | security-review |
Security-focused review with category coverage and exploit-path validation |
| Feature development | feature-dev |
Seven-phase workflow from discovery through implementation and review |
| Feature development | code-explorer |
Trace a feature across entry points, layers, state, and dependencies |
| Feature development | code-architect |
Produce a file-level architecture and implementation blueprint |
| Feature development | code-reviewer |
Adversarial review of a focused local change set |
| Design | frontend-design |
Create distinctive interfaces with an accessibility and craft rubric |
| Authoring | mcp-builder |
Design and build MCP servers in TypeScript or Python |
| Authoring | skill-creator |
Create, test, and improve reusable SKILL.md workflows |
| Project memory | agents-md-improver |
Audit project rules and propose targeted improvements |
| Project memory | agents-md-revise |
Capture durable session learnings in project rules |
code-explorer, code-architect, and code-reviewer are standalone skills and specialist roles used by feature-dev. OpenCode registers named least-privilege agents; Codex can carry out the same assignments with native subagents and can use matching custom agents when the user configures them.
Installation
Prerequisites
- Git
- One supported host:
- OpenCode 1.18.7 or newer: https://opencode.ai
- A current Codex CLI or Codex desktop environment with plugin support: https://developers.openai.com/codex/
Codex
Add this repository as a marketplace, then install the plugin:
codex plugin marketplace add waybarrios/opencode-power-pack --ref main
codex plugin add opencode-power-pack@opencode-power-pack
Start a new Codex session after installation so the eleven bundled skills are loaded. Use /plugins to inspect the installed plugin or $ to select one of its skills explicitly.
OpenCode From GitHub
opencode plugin --global "opencode-power-pack@git+https://github.com/waybarrios/opencode-power-pack.git"
Restart OpenCode after installation. OpenCode discovers each skill and creates its same-named slash command automatically.
To pin a published release, append its tag:
opencode plugin --global "opencode-power-pack@git+https://github.com/waybarrios/opencode-power-pack.git#<tag>"
OpenCode From A Local Clone
git clone https://github.com/waybarrios/opencode-power-pack.git ~/code/opencode-power-pack
opencode plugin --global "opencode-power-pack@git+file:///home/you/code/opencode-power-pack"
Use an absolute file:// URL adjusted for your operating system. The target directory must be a Git repository.
Verify OpenCode
opencode debug skill
opencode debug agent code-explorer
The first command should include all eleven unprefixed skill names. The second should report a subagent with editing denied. In the TUI, ctrl+p should list /code-review, /feature-dev, /frontend-design, and the other skill-derived commands.
Verify Codex
codex plugin list --marketplace opencode-power-pack
The list should show opencode-power-pack as installed. Start a new session and explicitly select $code-review, $feature-dev, or another bundled skill for a first test.
Updating
For Codex:
codex plugin marketplace upgrade opencode-power-pack
codex plugin add opencode-power-pack@opencode-power-pack
Start a new session after the upgrade.
For an OpenCode GitHub installation:
opencode plugin --global --force "opencode-power-pack@git+https://github.com/waybarrios/opencode-power-pack.git"
Restart OpenCode after the command finishes. For a pinned installation, update the tag first. Local-clone users should run git pull in the clone and reinstall with the same git+file:// spec.
Uninstalling
For Codex:
codex plugin remove opencode-power-pack@opencode-power-pack
For OpenCode, remove the opencode-power-pack@... entry from the plugin array in the config where it was installed, then restart OpenCode. There are no command symlinks or copied command files to remove.
Troubleshooting
| Symptom | Cause | Fix |
|---|---|---|
| Codex cannot find the marketplace | Marketplace snapshot is missing or stale | Run codex plugin marketplace add waybarrios/opencode-power-pack --ref main, or upgrade the existing marketplace |
| Codex installed the plugin but skills do not appear | The current session predates installation | Start a new Codex session and inspect /plugins |
| Skills or commands do not appear | OpenCode is older than 1.18.7 | Upgrade OpenCode, restart, and run opencode debug skill |
| Skills are missing from debug output | Plugin installation failed or its config entry is absent | Re-run opencode plugin --global --force <module-spec> and inspect the reported error |
| Specialist agents are missing | Stale plugin checkout | Force reinstall, restart, and run opencode debug agent code-explorer |
| Installation reports a Git error | Invalid URL, network failure, or inaccessible repository | Validate the source with git ls-remote <url> |
| A workflow is rushed or incomplete | The backing model skipped multi-stage instructions | Use a stronger model and inspect whether required subagent tools are available |
Invocation
Codex exposes installed skills through $ mentions, while OpenCode 1.18.7+ exposes each discovered skill as a same-named slash command:
/code-review
/security-review
/feature-dev
/code-explorer
/code-architect
/code-reviewer
/frontend-design
/mcp-builder
/skill-creator
/agents-md-improver
/agents-md-revise
Examples:
$code-review review the current branch against main
$feature-dev add a logout button to the topbar
/code-review --comment
/code-review review https://github.com/owner/repo/pull/449
/feature-dev add a logout button to the topbar
/security-review
/frontend-design pricing page, brutalist tone, single-screen
In OpenCode, an explicit command with the same name takes precedence over a skill-derived command.
How It Works
opencode-power-pack
|
+-- .codex-plugin/plugin.json
| +-- packages all skills/ for Codex and compatible plugin surfaces
|
+-- .agents/plugins/marketplace.json
| +-- exposes the repository as an installable Codex marketplace
|
+-- .opencode/plugins/opencode-power-pack.js
| +-- registers skills/ in config.skills.paths
| +-- registers code-explorer as a read-only subagent
| +-- registers code-architect as a read-only subagent
| +-- registers code-reviewer with read-only Git commands
|
+-- skills/<name>/SKILL.md
| +-- shared Codex/OpenCode workflow
| +-- host-native invocation metadata
| +-- immutable source and license metadata
|
+-- UPSTREAMS.json
+-- repository, commit, path, blob, date, and adaptation type
Each SKILL.md is the single source for its workflow. The OpenCode plugin derives specialist-agent prompts from those files at startup. Codex loads the same skills from the plugin and follows feature-dev's specialist assignments with its native subagent workflow; installing a skill does not itself create a named custom agent.
The packaged agents deny edits, external network access, and nested tasks. code-reviewer additionally allows a narrow set of read-only Git commands.
Scope And Non-Goals
| In scope | Out of scope |
|---|---|
| Portable Claude Code workflow methodology | Claude Code hook contracts |
| Codex and OpenCode skills, invocation, and subagent orchestration | Proprietary or non-redistributable plugins |
| Licensed adaptations with immutable provenance | Automatic trust of third-party skill catalogs |
| Explicit permission boundaries and regression tests | Supporting OpenCode versions older than 1.18.7 or obsolete Codex plugin formats |
Contributing
Contributions are welcome for:
- Improvements based on reproducible workflow failures
- Behavioral evaluations and adversarial fixtures
- Portable references and deterministic helper scripts
- New ports with a verified license and immutable upstream source
Before opening a PR:
npm test
npm run smoke:opencode
npm pack --dry-run
Live behavioral evaluation is opt-in, needs OPENCODE_EVAL_MODEL, and runs only from a source checkout:
OPENCODE_EVAL_MODEL=provider/model npm run eval:behavioral
See evals/behavioral/README.md for the replay, review, and acceptance workflow. Snapshots are content-addressed evidence from one reviewed model execution, not universal guarantees.
Every skill must use a lowercase hyphenated directory/name, provide a trigger-specific description, remain under 500 lines where practical, and record its exact upstream in UPSTREAMS.json.
Acknowledgments
The bundled skills are modified upstream works. This repository contributes their Codex and OpenCode adaptation, packaging, additional workflow guidance, tests, and compatibility layer.
| Upstream | Pinned source | Used for |
|---|---|---|
| Anthropic | claude-plugins-official@bdca23e8 |
Code review, feature development, frontend design, and project memory |
| Anthropic | skills@5128e186 |
MCP Builder and Skill Creator |
| Anthropic | claude-code-security-review@0c6a49f1 |
Security Review |
| Jesse Vincent | superpowers@6efe32c9 |
OpenCode plugin registration pattern |
See UPSTREAMS.json for exact source paths and Git blobs. See THIRD_PARTY_NOTICES.md for attribution and modification notices.
License
The wrapper code and original project material are MIT-licensed under LICENSE. Modified skills derived from Anthropic's official plugins and skills are Apache-2.0, except security-review, whose upstream is MIT. See THIRD_PARTY_NOTICES.md, UPSTREAMS.json, and LICENSES/ for exact terms.