.dotfiles

This skill should be used when the user needs to automate browser interactions, test UI workflows, verify page elements, take screenshots, upload files to forms, or interact with web applications during development and debugging. Use this skill for testing PDF upload dialogs, form interactions, dropdown selections, and visual verification of web pages.

Execute tasks from a tasks.yaml by orchestrating code-writer agents

Plan a feature using the prd-planner subagent

Break down a PRD into tasks following the task YAML spec

Manage tmux sessions, windows, and panes from the CLI — list state, read pane output, send keys/commands, create or kill sessions and windows, and inspect long-running terminal work. Use this skill whenever the user mentions tmux, asks to check what's running in another terminal, wants to start a background process in a tmux session, send a command to a running pane, capture or tail output from a tmux pane, attach context from a pane, or troubleshoot a stuck tmux process — even if they don't explicitly say "tmux command."

Plan a feature using the prd-planner subagent

Domain-Driven Design + Ports & Adapters (Hexagonal Architecture) implementation guide. Use this skill whenever the user invokes /ddd, asks to build a feature using DDD, wants to model a domain, design bounded contexts, implement hexagonal architecture, or review code for DDD compliance. Also trigger when the user mentions entities, value objects, aggregates, repositories, domain events, use cases, ports, adapters, or composition root in an architectural context.

Exa AI search API via CLI. Activate when user wants to search the web, find code examples, extract content from websites, get AI answers with sources, or perform automated research. Examples: "search for AI startups", "find rate limiter code in Go", "extract content from this URL", "research this topic", "find similar pages".

Explore a codebase to find opportunities for architectural improvement, focusing on making the codebase more testable by deepening shallow modules. Use when user wants to improve architecture, find refactoring opportunities, consolidate tightly-coupled modules, or make a codebase more AI-navigable.

Review a pull request, branch diff, or local changes with a proven step-by-step workflow focused on architecture and design fit, refactor opportunities, potential bugs, and missing test coverage. Use when asked to review code, review a PR, look for bugs, suggest refactors, or assess test gaps.

Text-based flow diagrams (Mermaid/ASCII) paired with a short written walkthrough to explain how code, architecture, or business logic works. Trigger on "/visuals", "diagram this", "visualize the flow", "show me how X works", "map out the architecture", "how do services interact", "what's the data model", "flowchart of", "sequence diagram", "state transitions", "explain the flow of", or any request to chart/map/draw how something moves through steps — even without naming a format. Also trigger for proposed architectures, pseudo-code, or verbal descriptions the user wants rendered as a diagram.

Research skill that delegates investigation to a single subagent. Use this skill whenever the user asks to research, investigate, explore, look into, or learn about any topic. Invoke it for questions like "research X", "what do we know about Y", "look into Z for me", "deep dive on", or "find out about".

Commit only the changes relevant to the current task or context. Never uses `git add -A` or `git add .` — always stages files selectively. Use when the user wants to commit work, save progress, or record changes to git.

Create a GitHub pull request for the current branch. Invokes /commit first to stage and commit any uncommitted changes, then uses the GitHub CLI (gh) to open a PR with a generated title and body. Use this skill whenever the user wants to open a PR, submit their work for review, push a branch and create a pull request, or says anything like "open a PR", "create a pull request", "submit for review", "push and PR", or "ship this".

Diagnose bugs, errors, and unexpected behavior — then suggest a fix. Use whenever the user describes something broken: error messages, crashes, wrong output, "this isn't working", "why is X happening". Delegates to a debugger subagent using a four-phase investigation (root cause → pattern → hypothesis → fix) before recommending anything. Never writes code — produces a structured prose diagnosis only.

Interview the user relentlessly about a plan or design until reaching shared understanding, resolving each branch of the decision tree. Use when user wants to stress-test a plan, get grilled on their design, or mentions "grill me".

Test-driven development with red-green-refactor loop. Use when user wants to build features or fix bugs using TDD, mentions "red-green-refactor", wants integration tests, or asks for test-first development.

Create distinctive, production-grade frontend interfaces with high design quality. Use this skill when the user asks to build web components, pages, or applications. Generates creative, polished code that avoids generic AI aesthetics.

Craft effective prompts for subagent delegation. Use this skill when delegating work to subagents to ensure clear, well-structured prompts that produce high-quality results.

Review and address all open PR comments — fetch, triage, implement changes, resolve, and summarize.

Rebase current branch onto target with conflict resolution

Manage task YAML files with status tracking and dependencies. Use this skill to verify, query, and update task status in tasks.yaml files.