Write and validate Architecture Decision Records following Shopware ADR conventions. Interactively creates ADRs with proper YAML front matter, structure selection (simple or multi-domain), and guided content. Validates existing ADRs against front matter rules, required coverage, structure, writing style, and Shopware-specific patterns. Use when creating new ADRs, writing architecture decisions, validating ADR files, checking ADR quality, or when user mentions "ADR", "architecture decision record", or "decision record".

Write a summary post about repository changes since a given date for Discord and Slack. Analyzes commits on main, groups by plugin, and produces separate platform-formatted posts.

This skill should be used when the user asks "how does X work?", "what's the architecture?", "help me understand the codebase", "find all components that use Y", "trace the data flow", mentions design patterns, component relationships, or is onboarding to an unfamiliar codebase. Routes architectural queries to ChunkHound for semantic code research; uses native Grep/Glob only for simple string or file pattern searches.

Generate conventional commit messages for the Shopware AI Coding Tools marketplace. Determines type, infers scope from plugin directory structure, and detects breaking changes. Use when generating commit messages in this repository.

Generate conventional commit messages for the Shopware core repository. Two modes: squash merge titles (title-only) for trunk merges, and full commit messages (title + body) for branch commits. Analyzes diffs, infers scope from Shopware's directory structure, and detects breaking changes. Use when the user explicitly asks to generate, write, or create a commit message, squash commit, commit title, or merge commit message. Do not activate during implementation work or when the user is writing code.

Draft a PR title (conventional commit format) and description (narrative prose with topical subsections) for PRs targeting non-trunk feature branches. Analyzes the diff against the target branch, detects related PRs in the chain, and asks targeted questions for missing context.

Generate Shopware-compliant PHPUnit migration tests that pass PHPStan and PHPUnit validation.

Generate Shopware-compliant PHPUnit unit tests that pass PHPStan and PHPUnit validation.

Peer-to-peer debate of PHPUnit test review findings within an Agent Teams wave. Receives own findings and peer findings, debates with co-reviewers via SendMessage, and outputs final stance with withdrawn findings and reasons.

Defense against adversary challenges in the team review defense round. Receives adversary challenges, engages each on its merits, and outputs defense stance with adopted, re-adopted, and withdrawn findings.

Reviews PHPUnit unit tests for quality and compliance. Validates test structure, naming conventions, attribute order, mocking strategy, and behavior-focused testing. Use when user requests "review test", "check test quality", "validate test", "analyze test compliance", or mentions reviewing Shopware unit tests.

Team-based PHPUnit test review with 3-5 independent reviewers and 1-2 devil's advocate agents reaching consensus through structured debate and adversarial red team challenge. Accepts flexible input (file paths, commits, branches, PRs, directories) and resolves to a list of test files. Each file is reviewed by 3 reviewers from a variable-size pool with balanced overlap. After round 1 consensus, advocates challenge findings and reviewers defend under adversarial rules. Cross-file consistency analysis identifies pattern divergences across files.

This skill should be used when the user asks to "write unit tests for", "generate tests for", "create PHPUnit tests", "add test coverage", "test this class", "cover this with tests", "I need tests for", "unit test this", "SW6 unit tests", "Shopware unit tests", "PHPUnit tests for Shopware", or mentions PHPUnit test generation for Shopware 6. Provides automated test generation with review-fix cycles that validate tests until they pass. Should NOT be used for integration tests, e2e tests, or non-PHP testing.

Update plugin versions and synchronize setup skills. Use when bumping a plugin version, updating a SETUP.md, or creating a new setup skill for a plugin. Handles version bumps across plugin.json, SKILL.md frontmatters, and setup skill synchronization from templates.

Draft a PR title (conventional commit format) and description (Shopware's 5-section template) by analyzing the full branch scope, leveraging session context, and asking targeted questions for missing context.

Draft entries for RELEASE_INFO and UPGRADE files in the Shopware core repository. Analyzes the full branch diff against trunk, synthesizes the narrative of the changes, and writes entries calibrated to the magnitude of change. Use when completing features, deprecations, or breaking changes that affect external developers, or when the user mentions writing release info, upgrade entries, release notes, release documentation, or changelog entries. Do not activate mid-implementation or for internal refactoring, non-critical bug fixes, or test-only changes.

Interactive setup for the chunkhound-integration plugin (semantic code research via ChunkHound). Checks that the chunkhound CLI and an embedding provider (VoyageAI, OpenAI, or Ollama) are available, creates .chunkhound.json with provider configuration, runs the initial index, and validates the MCP server connection. Use when the user installs chunkhound-integration and needs configuration, asks how to set up semantic search, or when ChunkHound MCP tools fail with config or connection errors.