Development

Turns a raw idea into an approved spec, including database schema and API-contract design. Use before planning, while intent is still open. For visual UI design, use design.

Writes the code for an already-defined task, test-first, following your codebase's own patterns. Also scoped fixes and performance optimizations on existing code.

Turns an approved spec into an ordered, dependency-sorted task plan with a shape decision per task. Implementation planning, not product roadmap planning.

Whole-app health report from a panel of experts — how the features connect, which are strong, which are weak, and what to improve, with market evidence. App-level, not diff-level.

The full software-engineering lifecycle as one suite — DISCOVER, DEFINE, PLAN, BUILD, VERIFY, REVIEW, SHIP — with approval gates before code, test-first builds, evidence before "done", and resumable multi-session work. Routes to 12 sub-skills covering any stack, plus feature discovery, security audit, UI/UX craft, and learning roadmaps.

Maintain ALL Heptabase comparison-overview card families with one skill — tokenizer/codec, spoken-LM/speech-LLM (nesting the auditory 聽覺理解・推理 sub-topic: audio understanding/reasoning/audio-LM 驗收), duplex S2S, 語音生成 (TTS family), ASR (non-LLM), and speech frontend/security. Find newly Tasks-tagged papers an overview is missing, route each to the right topic and act card, author a hung-yi-lee-style section + dimension row + paradigm/synthesis updates, re-sort by arxiv ID, refresh the 導讀, and kee...

Prevent AI coding agents from introducing architecture drift. Use before non-trivial features, refactors, cross-module fixes, new services, repositories, validators, state stores, dependencies, schema changes, public API changes, or edits governed by a change contract. Architecture Guardian provides bounded discovery, reuse-first search, dependency rules, reviewed baselines, short-lived change contracts, scoped expiring exceptions, deterministic reports, and optional strict gates.

Local-first governance, memory, runtime evidence, project isolation, architecture conformance, and release safety for AI coding agents. Use Agent Kernel when a user wants Claude Code, Codex, Cursor, OpenCode, Antigravity, Gemini CLI, or AGENTS.md-compatible agents to share persistent rules, reviewed workflows, Failure Lessons, episodes, bounded runtime context, project connection manifests, provider approvals, generated guidance, MCP tools, hooks, command guards, or architecture contracts. St...

Analizza, compone, implementa e verifica i ritorni a capo dei testi web in base a significato, punteggiatura e ritmo visivo. Utilizzare durante la creazione o revisione di siti, landing page, prototipi HTML e interfacce frontend quando hero, titoli, lead, card, CTA, navigazione o microcopy presentano interruzioni illogiche, parole orfane, righe sbilanciate, overflow o testi brevi che dovrebbero rimanere su una sola riga.

Evidence-aware maintenance for agent workflows: cleanup, retention, compaction, archival, and recovery of shared artifacts. Use when planning or carrying out workflow maintenance, investigating maintenance-related breakage, or evaluating whether maintenance reduces human supervision without damaging downstream work. Do not apply to unrelated coding tasks or use it to authorize business decisions. Works with ordinary scripts; WM integration is optional.

Writing guidelines for removing AI patterns and adding human voice. Use when writing, editing, or reviewing substantive content such as summaries and reports, or when the user asks to improve, simplify, or make text more concise.

Guards against generic AI design. Use when creating, styling, or reviewing any kind of visual design.

TypeScript best practices for clean, maintainable, and optimized code. MUST USE for writing or working with TypeScript code (.ts, .tsx, .mts files), including editing, reviewing or refactoring.

React code style, conventions, best practices, and patterns for building clean, efficient, and scalable applications. MUST USE for writing or working with any React code, components, hooks, or JSX (.jsx, .tsx files), including editing, reviewing or refactoring.

Installs, upgrades, and removes Node.js/npm dependencies, runs scripts, executes packages, and manages package workflows. MUST USE for any task involving npm, npx, pnpm, yarn, or bun, including package.json dependency changes, lockfile updates, install/update/uninstall actions, and script execution.

Turn any codebase, product, or blog post into a high-retention 9:16 vertical viral reel using Hyperframes. Trigger with "/reel", "/reelcraft", "make a reel for this", "create a short video", or "generate a viral tiktok".

Use when translating brand strategy into visual creative direction, such as writing a creative or logo brief, describing mood boards, or setting photography, illustration, color, and type direction for a designer. Keywords - visual identity, visual direction, creative brief, mood board, logo brief, photography style, Alina Wheeler, visual language, design direction.

Use when creating, shortlisting, or testing taglines, slogans, or brand mantras, or distilling a positioning statement into a few memorable words. Keywords - tagline, slogan, catchphrase, brand motto, Neumeier trueline vs tagline, Ogilvy headline principles, Schwartz desire channeling, distillation method, tagline testing.

Use when drafting, editing, or reviewing flat, robotic, disconnected, repetitive, performative, overwritten, or overly harsh prose, especially in emails, feedback, criticism, and negotiation.

Use when researching or defining a target audience for brand work, building customer profiles from motivations rather than demographics, or explaining why customers switch. Keywords - Jobs to be Done, JTBD, functional emotional and social jobs, four forces, push pull anxiety habit, psychographics, limbic types, VALS segments, customer motivations.

Align with the user, then design the smallest useful managed AI work graph for an objective involving multiple steps, sources, parallel paths, checks, risk, or approvals. Produces an approval-ready graph with evidence requirements, validation routes, and human gates, and never executes its jobs. Not for knowledge graphs or simple one-pass tasks.

Use when writing or coaching a spoken brand pitch, such as an elevator pitch, one-liner, founder intro, or 10, 30, and 60 second variations, or when a pitch sounds written rather than spoken. Keywords - elevator pitch, one-liner, verbal pitch, StoryBrand SB7, Duarte Sparkline, Westfall CLARITY, Raskin strategic narrative, Golden Circle.

Use when analyzing competitor brands to find visual or positioning white space before making identity decisions, such as auditing competitor colors, typography, logos, imagery, and voice, or building a perceptual map. Keywords - competitive audit, competitor analysis, visual landscape, differentiation, white space, perceptual mapping, zig when others zag.

Use when defining brand voice, tone of voice, or verbal identity, writing voice guidelines, or adapting tone across channels and situations. Keywords - brand personality traits, BrandSort, Margot Bloomstein, Nielsen Norman four dimensions of tone, Aaker brand personality, this but not that, tone matrix, vocabulary do and don't lists.