cc

MUST BE LOADED for any coding task: implementing features, fixing bugs, writing code, refactoring, or making changes. This skill provides the mandatory step-by-step SDLC (Software Development Lifecycle) workflow for orchestrating development using specialized skills and sub-agents. Load this skill when the user asks to 'add', 'create', 'build', 'fix', 'update', 'change', 'implement', or 'refactor' anything.

MUST BE LOADED BEFORE modifying docs/tasks.md, docs/changes/, or task lists. Also load when user mentions tasks.md, changes/, 'the task file', specs, or discusses tasks/tracking. Triggers include: 'add a task', 'update tasks', 'mark as done', 'mark complete', 'track this', 'next task', 'what's next', 'work on next', create tickets/issues, 'add a feature', 'improve X to allow Y', plan features, break down tasks, or ANY project/task management discussion. This skill handles all task tracking through docs/tasks.md, docs/changes/ task lists, or external tools (GitHub Issues, Jira).

Spawn a coordinated sub-agent team to implement a spec or multi-task feature in parallel. Use when user says '/team', 'use a team', 'parallel implementation', 'work on multiple tasks', or provides a spec/issue that requires coordinated multi-PR work. The main session acts as coordinator — no code, no commits, only delegation and quality control.

Write, update, and maintain living specification documents and propose change documents. Use when user says 'write a spec', 'create a spec', 'spec out', 'update spec', 'spec this', 'design a feature', 'write a feature spec', or needs to create/modify specs in docs/specs/ or propose changes in docs/changes/. Handles the complete spec lifecycle: creation, updates, gap analysis, and change proposals.

MUST BE USED when user asks to: implement a feature, fix a bug, write code, add functionality, build something, code this, make changes. Implements code changes — when used within the SDLC workflow, PR creation is handled separately by pr-preparer.

Run OpenAI Codex's AI code review LOCALLY via the `codex` CLI BEFORE opening a PR (part of pre-PR self-review). Runs one-shot, reviewing the current branch against main, prints findings to stdout, and you resolve them before the PR is opened. Use during pre-PR self-review, alongside fx-dev:coderabbit-review.

Process and resolve GitHub Copilot automated PR review comments. Use when the user says "check copilot review", "handle copilot comments", "resolve copilot feedback", "address copilot suggestions", or mentions Copilot PR comments. Also use after PR creation when Copilot has left automated review comments.

MUST BE USED when working with GitHub: updating PRs, editing PR descriptions/titles, creating PRs, merging, review threads, `gh` CLI commands, GitHub API, or any pull request operations. Load this skill BEFORE running gh commands or modifying PRs. (plugin:fx-dev@fx-cc)

MUST BE USED when user asks to: plan a feature, create a plan, break down a task, design architecture, figure out how to implement. Creates detailed implementation plans by breaking down requirements into actionable steps.

MUST BE USED when user asks to: open a PR, create a PR, submit a PR, make a PR, prepare a PR, push changes for review. Prepares pull requests by analyzing branch changes, reviewing commits, creating the PR on GitHub, and ensuring compliance with project standards.

MUST BE USED when user asks to: review code, review PR, check my code, look at my changes, review changes. Reviews pull requests and code changes, evaluating quality and providing actionable feedback.

Process and resolve CodeRabbit automated PR review comments. Use when the user says "check rabbit review", "handle coderabbit comments", "resolve rabbit feedback", or mentions CodeRabbit PR comments. Also use after PR creation when CodeRabbit has left automated review comments.

Fetch and analyze GitHub issues, extract requirements, gather context from referenced URLs, and compile comprehensive requirements documentation for implementation.

Process Codecov coverage reports on PRs and add missing tests for uncovered lines. Use when Codecov reports insufficient coverage, the user says "check codecov", "fix coverage", "add missing tests", or after CI passes and Codecov posts its report. Also invoked automatically by the SDLC workflow and resolve-pr-feedback skill.

Check a PR for unresolved automated review feedback (Copilot, CodeRabbit, Codecov) and invoke the appropriate resolver skills. Use when the user says "resolve PR feedback", "check PR comments", "address review comments", "fix coverage", or wants to handle all automated review feedback on a PR.

Initialize the docs/ folder structure for spec-driven development and create the standard AI instruction files (AGENTS.md, REVIEW.md, the CLAUDE.md pointer, and .coderabbit.yaml) if missing, so task tracking defers to /project-management. Creates defaults only — never migrates existing files; that is fx-dev:upgrade. Called automatically by /spec-writer and /project-management. Also use when user says 'setup docs', 'initialize docs', 'create docs structure'.

Migrate a repo's AI INSTRUCTION FILES to current fx-dev conventions — moving CLAUDE.md content into AGENTS.md, folding .github/copilot-instructions.md into REVIEW.md, resolving instruction-file symlinks. Intentionally intrusive; rewrites and deletes those files. Use ONLY for instruction-file/convention migration: 'migrate conventions', 'update instruction files', 'upgrade fx-dev conventions', 'migrate to AGENTS.md', or when fx-dev:setup reports a legacy layout. NOT for upgrading dependencies, packages, frameworks, language versions, or databases — those are ordinary code work, use fx-dev:dev.

Contribute local component changes upstream to fx/ui. This skill should be used when the user wants to upstream local UI component changes (variants, new components, fixes) from a consumer repo to the shared @fx/ui library. Triggers include "upstream to fx/ui", "contribute to fx/ui", "add this to fx/ui", "move this to fx/ui", "send PR to fx/ui", or when local UI component changes should be shared across repos.

Update fx-cc plugin skills based on conversation learnings. Use when the user says "use /learn to...", "learn to...", "remember to...", "don't do X again", or when a skill misbehaved and needs correction. This skill modifies plugin source files but does NOT commit changes - they require manual review before committing.

Run CodeRabbit's optional AI review. PRIMARY path: run it LOCALLY via the `cr` CLI before opening a PR and resolve actionable findings. FALLBACK path: wait for + resolve its automated PR review when available. Rate limits degrade gracefully: report once, skip CodeRabbit, and continue the SDLC.

MUST BE USED when: PR checks are failing, CI is red, tests failing on PR, build failed, need to monitor PR status. Monitors GitHub pull request checks and coordinates fixes for failures.

Analyze and fix CI check failures on a PR. Use when CI checks fail, build errors occur, or tests fail in CI. Invoked automatically by the SDLC workflow (Step 7.4) when checks fail, or manually via 'resolve CI failures', 'fix CI', 'fix failing checks'.

Request, wait for, and resolve GitHub Copilot's automated PR review. Use after creating or marking a PR ready. Copilot reviews EVERY PR automatically — this skill handles the full lifecycle: request review, poll until received, then resolve all feedback. MUST be used before merging any PR.

MUST BE USED when user asks to research a project, validate a product idea, scope a new application, evaluate market viability, or lay out the foundational approach for building a full application. Triggers: 'research a project', 'research project', 'should we build X', 'is X viable', 'lay out the approach for X', 'scope out X', 'validate this idea', 'competitive analysis', 'market analysis', 'how should we build X'. Performs the most thorough possible end-to-end project research: competitive landscape discovery, deep competitor analysis, market viability assessment, constraint validation, and at least two fully-reasoned solution proposals — followed by adversarial self-validation via a dedicated red-team sub-agent.

MUST BE USED for bug fixes that require test-first verification. Use when user says '/fix', 'fix this bug', 'fix the error', 'debug this', or provides an error/bug report. This skill mandates writing a failing test BEFORE implementing any fix, then loads the SDLC workflow.

Verify web application changes work by launching the app stack and testing in a real browser. This skill should be used when the user asks to "verify the change", "test in browser", "check if it works", or after completing a PR to validate the implementation. Requires Playwright MCP server. MUST exit if Playwright MCP is unavailable.

Updates GitHub issues with planning info, status changes, and progress, ensuring proper tagging and documentation throughout development.

Reviews pull requests or changesets to ensure only minimal, necessary changes are included, with no extraneous modifications or artifacts.

MUST BE USED proactively to execute complete workflows from start to finish without stopping. Proactively ensures all phases complete and loops until success.

Guide for creating effective fx/cc marketplace plugins. This skill should be used when users want to create a new Claude Code plugin (or update an existing plugin) for the fx/cc marketplace that extends Claude's capabilities with skills, commands, or hooks.

Research and recommend libraries, technologies, or software solutions for specific use cases or project requirements.

This skill should be used when working with Steam on Linux - managing non-Steam game shortcuts, configuring Proton/Wine compatibility, parsing VDF files, or finding Steam paths and prefixes. MUST BE USED when user mentions Steam launch options, modifying shortcuts.vdf, Proton environment variables (PROTON_*, WINE*), Battle.net/Blizzard games on Linux, or RTX 4000+ performance fixes. ALSO USE when troubleshooting Proton/Wine game issues (black screen, game won't launch, launcher issues, CEF/Chromium rendering problems).

Updates GitHub Release notes with human-readable summaries AFTER release-please creates a release. Use when user says "update release description", "improve release notes", or after merging a release-please PR. NEVER modifies the release PR body (breaks parsing).

MUST BE USED when user says: 'create a skill', 'make a skill', 'new skill', 'build a skill', 'write a skill', 'add a skill', 'skill for X', or any variation. Also use when editing SKILL.md files, updating skill descriptions, adding skill resources, or iterating on existing skills. Load this skill BEFORE any skill-related work.

Provides comprehensive guidance for managing MCP (Model Context Protocol) servers using the claude mcp CLI. Covers adding servers across scopes (project, local, user), transport types (stdio, sse, http), security best practices, and common patterns. Strongly recommends project scope for team consistency and version control.