skill-issue
Run AI agent skills as CLIs to orchestrate parallel code reviews across Claude and OpenAI on git diffs/PRs, synthesize findings, generate/verify plans as Linear issues, simplify code changes, define testing strategies for Swift/Python/TS, and check architecture/security.
npx claudepluginhub crown-dev-studios/skill-issue --plugin skill-issueComponent Overview
Component Details
Skills (11)
Review a plan or implementation for model integrity, service boundaries, and canonical architecture direction. Use when a change touches service boundaries, data models, or trust boundaries.
Clarify what should be built before planning begins. Use when the problem is under-defined, the intended outcome is unclear, or there are materially different directions to explore before committing to a plan.
Directly simplify existing code changes for clarity and minimal complexity. Use after writing or modifying code when the user wants unnecessary complexity, defensive patterns, over-abstraction, excess state, or excess line count removed by making edits. Applies to any language. Triggers on direct requests to simplify code, reduce complexity, clean up changes, or make code more elegant. Supports an explicit manual `mode:review` argument for findings-only simplicity review.
Convert a plan of record into Linear-ready issues with dependencies, acceptance criteria, and sequencing. Use when a reviewed plan needs to become executable work in Linear.
Check whether an implementation matches its plan of record. Use when a plan exists and code has been written against it. Flags deviations, missing phases, unimplemented acceptance criteria, and scope drift.
Challenge and strengthen a plan or brainstorm before implementation begins. Use when a plan of record or brainstorm document exists and needs a second pass for scope, sequencing, testing, complexity, or operability gaps.
Create a concrete plan of record from a clarified idea, existing code, or a spec. Use when a task crosses module boundaries, touches service boundaries, or requires choosing between viable approaches.
Orchestrate model-parallel code review with selected skill references. Runs Claude and Codex in parallel with those skill references listed in reviewer prompts as additional review lenses, then synthesizes all findings through an LLM judge with semantic deduplication, contradiction detection, and dependency ordering.
Classify and route review feedback before implementation begins. Use when review findings exist from any source and need validation, prioritization, and routing to the right next action.
Get a second opinion on your current conversation from a different AI model. Routes to Claude from Codex and Codex from Claude using a fresh local shell session.
Testing philosophy and conventions. Use when writing, generating, modifying, or auditing test files. Covers behavior-driven testing, async patterns, database testing, and language-specific practices for Swift, Python, and TypeScript.
README
skill-issue
First-party AI agent skills and companion CLIs for software development workflows. Built and maintained by Crown Dev Studios.
Install
Claude Code
Add the marketplace and install the plugin:
/plugin marketplace add crown-dev-studios/skill-issue
/plugin install skill-issue@skill-issue
All 11 skills are declared through .claude-plugin/plugin.json and activate immediately.
Codex, Cursor, and other ~/.agents/skills-aware harnesses
Run the install script — it copies each skill into ~/.agents/skills/skill-issue/, stripping git metadata, tests, and build sources:
curl -fsSL https://raw.githubusercontent.com/crown-dev-studios/skill-issue/main/scripts/install.sh | bash
Or from a local checkout:
./scripts/install.sh
Override the destination with SKILL_ISSUE_DEST=/some/other/path ./scripts/install.sh.
npm (CLIs only)
The repo also ships as one npm package with multiple commands:
npx @crown-dev-studios/skill-issue second-opinion --help
npx @crown-dev-studios/skill-issue review-council --help
If you install it globally, you also get direct command names:
npm install -g @crown-dev-studios/skill-issue
skill-issue --help
second-opinion --help
review-council --help
skill-issue is the canonical release unit. second-opinion and review-council are published together from the same root package so their versions stay in sync.
What's in here
These are the first-party skills we use to keep our engineering standards high and our feedback loops tight.
| Skill | What it does |
|---|---|
| architecture-review | Reviews plans or implementations for model integrity, service boundaries, and canonical architecture direction using SOLID principles as a lens. |
| brainstorming | Clarifies what should be built before planning begins. Resolves ambiguity in the problem, outcome, or direction through structured interview and option exploration. |
| plan-review | Challenges and strengthens plans or brainstorms before implementation. Reviews for scope, product framing, sequencing, complexity, testing, operability, error handling, and threat model. |
| planning | Creates a plan of record that serves as both spec and execution plan. Covers current state, constraints, invariants, model and API boundaries, architecture diagrams, phased execution, and proof strategy. |
| testing-philosophy | Enforces our testing principles: what to test, how to structure tests, and when to push back on coverage theater. |
| second-opinion | Asks a different AI agent for a second take on the current thread. Routes to Claude from Codex and Codex from Claude. |
| linear-issue-shaping | Converts plans of record into Linear issues with dependencies, milestones, acceptance criteria, and sequencing. |
| review-council | Runs parallel code reviews, then synthesizes and ranks the feedback to surface what actually matters. |
| review-triage | Classifies and routes review feedback before implementation. Validates findings, assigns severity (P1/P2/P3), and routes to fix now, follow-up ticket, follow-up plan, or dismiss. |
Development
From the repo root:
pnpm install
pnpm run build
pnpm run test
pnpm run pack:dry-run
Local command entrypoints are also exposed as root scripts:
pnpm run second-opinion -- --help
pnpm run review-council -- --help
Packaging and release
The root VERSION file and package.json are the single source of truth. Release scripts live in scripts/ and are exposed as pnpm run release:*.
Preflight (no publish) — install with frozen lockfile, build, test, and dry-run npm pack:
pnpm run release:preflight
Bump the version (updates VERSION, package.json, lockfile metadata, then commits and tags vX.Y.Z):
bash scripts/bump-version.sh patch # or minor | major | 1.2.3
bash scripts/bump-version.sh patch --no-push
Verify the tag matches VERSION and package.json:
bash scripts/check-version.sh --require-tag
Publish to npm (runs preflight, pushes branch + tag, then pnpm publish). Requires npm login, a clean git tree, and the matching vX.Y.Z tag:
bash scripts/deploy.sh --dry-run # preview
bash scripts/deploy.sh # publish
The pnpm run release:* aliases invoke the same scripts but do not forward extra flags cleanly, so call the scripts directly when passing options. pnpm run release:preflight remains the convenient no-arg entry point for the preflight.
Full release sequence:
bash scripts/bump-version.sh patch
bash scripts/check-version.sh --require-tag
bash scripts/deploy.sh
License
MIT
Similar Plugins
chrome-devtools-mcp
Reliable automation, in-depth debugging, and performance analysis in Chrome using Chrome DevTools and Puppeteer
ChromeDevToolsgodot-claude-skills
Claude Code skills for Godot 4.x game development - GDScript patterns, interactive MCP workflows, scene design, and shaders
planning-with-files
Manus-style persistent markdown files for planning, progress tracking, and knowledge storage. Works with Claude Code, Kiro, Clawd CLI, Gemini CLI, Cursor, Continue, Hermes, and 17+ AI coding assistants. Now with Arabic, German, Spanish, and Chinese (Simplified & Traditional) support.
gsd-build-get-shit-done
Meta-prompting and spec-driven development system for Claude Code. Productivity framework for structured AI-assisted development.
superpowers
Core skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques
Help us improve
Share bugs, ideas, or general feedback.
