dallenpyrah

Apply Emil Kowalski-style UI polish: interaction details, component feel, animation restraint, and invisible product quality. Uses progressive disclosure instead of long design essays.

Apply Emil Kowalski-style animation guidance for web interfaces: purposeful motion, correct easing, interruptibility, performance, reduced motion, and interaction feel.

Implement the requested change on the current branch using the best available contract: issue, architecture, user request, or repo evidence. Does not require /architect or /review to have run. Hands naturally to /test.

Triage and address PR review comments. Uses available issue/architecture/context when present, keeps evidence in .context/address.md, and does not require prior workflow phases.

Derive the smallest correct architecture from user intent, repo facts, and available workflow evidence. Writes the full architecture to .context/architecture.md and returns only the decision, tradeoff, risk, and next action.

Review the current PR for correctness, tests, maintainability, security, standards, and prior-findings risk. Keeps reviewer detail in .context/code-review.md and posts concise actionable GitHub review comments.

Reproduce, root-cause, and fix a bug with the smallest verified change. Writes investigation evidence to .context/debug.md and hands naturally to /test.

Update documentation for changed behavior, APIs, config, operations, migrations, or architectural contracts. Reads current diff and evidence artifacts, writes only necessary docs, and hands naturally to /pr.

Work through review comments on the current PR. Triages each comment as Address / Push-back / Escalate and executes immediately — no approval gate. Addressed comments produce code changes, commits, and silent thread resolution via GraphQL (no reply text). Pushed-back comments stay unresolved. Escalations invoke /interview or /architect inline. Watches GitHub Actions after push — the loop is not complete until all checks pass.

Re-derive the simplest, cleanest architecture from first principles. Enforces minimal code, composition, single source of truth, deep modules, clear abstractions, clean interfaces, functional core, state machines for lifecycle, ports/adapters, typed recovery, no silent fallbacks, Effect primitives in Effect-owned code, and game-theoretic incentive compatibility. Output uses ASCII diagrams and code contracts, never mermaid, and fits one to two screens. Hands off to /review.

Fan out six focused reviewer personas against the current PR — correctness, testing, maintainability, project-standards, security, and previous-findings. Each reviewer applies first-principles and game-theoretic pressure: invariants, incentives, information asymmetry, local shortcuts, repeated-game health, and adversarial behavior. Dedupe, validate, verify diff-line commentability, then post one GitHub review with summary plus inline comments where possible. No autofix, no approval, no request-changes. Hands off to /address.

This skill encodes Emil Kowalski's philosophy on UI polish, component design, animation decisions, and the invisible details that make software feel great.

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.

Interview the user in five-question rounds until the problem, values, constraints, non-goals, incentives, failure modes, and solution direction are locked. Uses first-principles decomposition, game-theoretic incentive analysis, concrete examples, 5 Whys, laddering, QFT-style question selection, and Socratic pressure. Hands off to /architect.

Open a GitHub issue (or epic) from the locked architecture using a Linear + Shape Up + Larson-synthesized structure. Header (Owner · Appetite · Status · Part of), Problem, Outcome, Solution sketch with ASCII diagrams and code contracts, Files, Acceptance criteria, Rabbit holes, No-gos, References, Depends on. No mermaid. Hands off to /work.

Capture the post-mortem for a compound-engineering cycle. Reads the issue, PR, review comments, git log, and final diff. Extracts what worked, what changed, what incentives caused friction, what mechanisms improved code health, and what durable rule should be considered. Writes docs/learnings/YYYY-MM-DD-slug.md, commits it, pushes the branch, and hands off to /merge when the PR is still open.

Searches code, docs, symbols, git history, and implementation concepts by meaning with the semantic_search tool. Use when locating where behavior is implemented, tracing a feature across files, finding callers/usages by description, or asking why/when code changed.

Find code, docs, and symbols by meaning across the whole project. Use when you need to locate where something is implemented, trace a feature, behavior, or concept across files, understand an unfamiliar codebase, find all callers or usages, search by description ("where do we handle retries", "find the auth middleware"), or ask why/when something changed. Use the single semantic_search tool instead of multiple grep and read calls — it returns ranked file-and-line snippets in one call and uses far less context.

Run scheduled routines on the user's machine via the local `routines` daemon (Effect Cluster backed). Use when the user wants to create, list, run, inspect, or start/stop local scheduled routines (cron or interval) — e.g. "set up a morning check routine", "run the morning-check routine now", "start the routines daemon", "show recent routine history". Six tools cover init, list, run, history, daemon start, and daemon stop.