Answer questions about how something works in this codebase by reading the actual source code. Use when the user asks "how does X work", "where is X defined", or "what happens when X" and the answer requires reading multiple files.
Analyse uncommitted branch changes, propose a logical commit breakdown, and create the commits after user approval. Use when changes are ready to be committed.
Generate a feature architecture document by reading the code on the current branch, then propose lessons learned for codification. Use after implementing a feature to create Mermaid diagrams, data flows, deployment steps, and distil reusable patterns into project rules.
Implement a feature or change following established project patterns. Use when you have a clear plan or task description and want code written across backend and frontend layers following project conventions.
Plan feature implementation by combining existing patterns with a capability-first strategy for novel features, producing a file-by-file plan. Use before writing code for features that span multiple layers.
Debug frontend/UI bugs using runtime evidence. Starts a log server, instruments code, collects logs, and fixes based on evidence. When Playwright MCP is available, automates bug reproduction and verification via browser automation instead of asking the user to manually interact. Use when investigating UI bugs, state…
Analyse the impact of a proposed change across backend and frontend layers. Use when planning a modification to understand what files, types, and integrations will be affected before writing code.
Run frontend and backend preflight checks (lint, types, build, a11y, runtime smoke test via Playwright, PHP quality) and report results. Use before committing or pushing to verify the codebase is clean.
Review code changes (current branch, PR, or specific files) for correctness, patterns compliance, and cross-boundary consistency. Use after making changes or before merging.
Runs before the agent uses a tool for Write, Edit and Bash tool calls, executing core-vendor-guard.sh, clean-build-output.sh, graphql-sync-check.sh and validate-commit-msg.sh via bash (4 commands). From damian-demasi-aligent/ai-agent-tool-chain-generic.
Runs after a tool call finishes for Write and Edit tool calls, executing sync-manuals-check.sh and react-lint-on-edit.sh via bash (2 commands). From damian-demasi-aligent/ai-agent-tool-chain-generic.
Save critical session state to disk before running /compact. Captures branch, ticket, workflow phase, plan status, modified files, and verification results so context can be restored after compaction.
Divide all uncommitted branch changes into logical, layered commits following the project's commit conventions. Spawns the committer agent to analyse and propose a plan, presents it for user approval, then executes.
Amend an implementation plan mid-feature when assumptions prove wrong, requirements change, or the reviewer flags a fundamental issue. Compares the plan to current state, proposes targeted amendments, and updates the plan file.
This skill should be used when debugging frontend/UI bugs that need runtime evidence. USE THIS SKILL (instead of adding console.log) when you're about to: "add console.log and ask user to check", "open DevTools and tell me what you see", "reproduce the bug and share the output", "check the browser console". Triggers…
Auto-detect the project's technology stack and generate a stack configuration file. Run this before /setup-project when adopting the toolchain into a new project.
Generate a feature architecture document for a completed feature. Creates Mermaid diagrams, data flows, deployment steps, and captures screenshots via Playwright when available. Use after implementing a feature.
Magento 2 transactional email patterns. Use when creating or modifying email templates, TransportBuilder calls, emailtemplates.xml, or any feature that sends transactional emails.