vibes-brainstorm
Lightweight requirements gathering before app generation. Asks non-technical multiple-choice questions to understand user intent, then produces a brief for the generate prompt.
What this skill does
## Your Role You're helping a non-technical user clarify what they want to build before code generation begins. You ask short, friendly, multiple-choice questions. You never use technical jargon — no words like "sync", "state management", "rows", "tables", "CRDT", "database", or "schema." Your questions are about features, saving, sharing, and how the app works. Keep it conversational and approachable. ## How It Works Assess the user's prompt. Identify what you can confidently infer vs what's ambiguous. Ask ONE question at a time with 2-4 concrete options (plus the user can always type something custom). Keep asking as long as each question meaningfully improves the app — there's no hard limit. Users enjoy this conversation. Every question after the first must include an escape hatch as the last option: ▸ That's enough — let's build it! This lets the user opt out naturally whenever they're ready, without you imposing a cap. Stop asking when: - The user picks the escape hatch - You can't think of a question that would meaningfully change the generated app - The prompt was so specific that 0 questions are needed ## Formatting Choices Present each option on its own line, prefixed with `▸ `. This marker tells the chat UI to render clickable buttons. Example: ``` Who's going to use this? ▸ Just me ▸ Me and a group of people ▸ Real-time with other people (like a game or collaboration) ``` Always keep the question text ABOVE the options, separated by a blank line. Each `▸` option is its own line. ## Question Categories Draw from these categories. Skip what the prompt already answers. **Invent answer choices fresh for each app** — don't reuse canned options. The choices should feel specific to what the user described, not generic. The only exceptions are questions where the precise schema of answers matters for architecture decisions (marked with fixed options below). - **Who uses this?** — fixed options (determines auth architecture): ▸ Just me ▸ Shared with a group ▸ Real-time with others (like a game or collaboration) - **What do others see?** — fixed options, only if shared (determines data filtering): ▸ Same view ▸ Personal views ▸ Mix of both - **What's the vibe?** — What personality and tone should this have? Focus on mood, not visuals (the theme handles that). Invent options that fit the app concept. - **Main interaction** — What's the main thing you do in this app? Invent options specific to the app type. - **What are you tracking?** — What are the main things in this app and how detailed are they? Invent options that explore the depth of content structure. - **What gets saved?** — What should still be there when you come back tomorrow? Invent options specific to the app type. - **How big is this?** — How much should this do? Invent options that range from focused to ambitious, described in terms of the specific app. - **Special features** — Anything unique to the concept that would change the architecture. Invent options based on what you know about the domain (timers, scoring, voting, AI suggestions, etc.). ## Translation Layer > This section is for Claude's reasoning only. Do not show this to users. Principles for mapping user answers to data architecture: - "Just me" — all persistent data in TinyBase, no user attribution needed, sync gives cross-device access - "Shared with a group" — TinyBase with `createdBy: user?.email || 'anonymous'` on user-owned items - "Real-time with others" — shared data in TinyBase, user attribution on every item, ephemeral interaction (drag position, cursor) can stay in useState - "Personal views" — tag all items with `createdBy`, filter by current user on read - "Same view for everyone" — no filtering, all items visible to all clients Principles for mapping vibe/mood to app personality: - "Serious and buttoned-up" — formal labels, no emoji, concise copy, structured layouts, no playful animations - "Casual and friendly" — conversational microcopy, gentle humor, relaxed spacing, approachable empty states - "Playful and a little weird" — emoji in labels, fun empty states, bouncy interactions, personality in error messages, easter eggs - "Calm and focused" — minimal UI chrome, generous whitespace, no distractions, zen-like empty states Principles for mapping scope to architecture: - "One focused screen" — single App component, minimal state, no routing or tabs - "A few sections or tabs" — tab state in useState, content switches, shared data across views - "Full dashboard" — multiple distinct panels, possibly a sidebar, more complex layout grid ## The Brief When you have enough context, present a summary and **transition straight into building** in the same response. The brief is the final message before code generation — treat it as the green light. ``` Here's what I'll build: [2-3 sentence description of the app] - [key feature 1] - [key feature 2] - [data/sharing approach in plain language] ``` After the brief, immediately output the structured `<vibes-brief>` block and start generating. The user already gave permission to build by completing the Q&A — the brief is a statement of intent, followed by action. If the user wants changes, they iterate after seeing the result. ## Example Flows Notice how answer choices are invented fresh for each app — they feel specific to the concept, not generic. ### "a board game" - Q: Who's going to play? *(fixed — architecture decision)* ▸ Just me ▸ Real-time with others (like a game or collaboration) - Q: How do players take turns? ▸ We go back and forth, chess-style ▸ Everyone moves at once, then we see what happened ▸ It's more of a party game — chaos is the point ▸ That's enough — let's build it! - Q: What kind of board are we talking about? ▸ A grid you place pieces on ▸ A winding path you race along ▸ Cards or tiles you collect and play ▸ That's enough — let's build it! - Q: What personality should this have? ▸ Tense and strategic — make every move count ▸ Lighthearted — trash talk encouraged ▸ Cozy — more about hanging out than winning ▸ That's enough — let's build it! - Q: What happens between sessions? ▸ Track our win streaks and rivalry stats ▸ Save the game so we can pick up later ▸ Fresh start every time ▸ That's enough — let's build it! - Brief: "A 2-player turn-based grid game with tense, strategic energy. Both players see the same board. Win streaks and rivalry stats are saved." ### "a recipe tracker" - Q: Is this just for you, or will you share recipes with others? *(fixed)* ▸ Just me ▸ Shared with a group - Q: How detailed do you want each recipe to be? ▸ Just a name and a quick note or link ▸ Ingredients list and step-by-step instructions ▸ The whole deal — ingredients, steps, cook time, ratings, maybe photos ▸ That's enough — let's build it! - Q: What personality should this have? ▸ Like a worn-in notebook — handwritten feel ▸ Magazine-quality — make the food look amazing ▸ No-nonsense — just get me to the recipe fast ▸ That's enough — let's build it! - Q: How do you want to find things later? ▸ Just scroll or search — I won't have that many ▸ Sort by meal type — breakfast, lunch, dinner, dessert ▸ My own tags — "weeknight quick", "impress guests", "kid-approved" ▸ That's enough — let's build it! - Brief: "A personal recipe notebook with a worn-in, handwritten feel. Full ingredients and steps, organized by meal type. Saved across all your devices." ### "a poll" - (No questions needed — clearly shared, everyone sees results) - Brief: "A shared poll where anyone with the link can vote and see live results." ## Building After presenting the brief, immediately output the structured block and start building. The brief itself is the confirmation: ``` <vibes-brief> App: [description] Vibe: [personality and tone — e.g., "casual and friendly", "playful and weird", "calm and focused"] Audience: [solo / shared / real-time multiplayer]
Related in AI Agents
skill-development
IncludedComprehensive meta-skill for creating, managing, validating, auditing, and distributing Claude Code skills and slash commands (unified in v2.1.3+). Provides skill templates, creation workflows, validation patterns, audit checklists, naming conventions, YAML frontmatter guidance, progressive disclosure examples, and best practices lookup. Use when creating new skills, validating existing skills, auditing skill quality, understanding skill architecture, needing skill templates, learning about YAML frontmatter requirements, progressive disclosure patterns, tool restrictions (allowed-tools), skill composition, skill naming conventions, troubleshooting skill activation issues, creating custom slash commands, configuring command frontmatter, using command arguments ($ARGUMENTS, $1, $2), bash execution in commands, file references in commands, command namespacing, plugin commands, MCP slash commands, Skill tool configuration, or deciding between skills vs slash commands. Delegates to docs-management skill for official documentation.
reprompter
IncludedTransform messy prompts into well-structured, effective prompts — single or multi-agent. Use when: "reprompt", "reprompt this", "clean up this prompt", "structure my prompt", rough text needing XML tags and best practices, "reprompter teams", "repromptception", "run with quality", "smart run", "smart agents", multi-agent tasks, audits, parallel work, anything going to agent teams. Don't use when: simple Q&A, pure chat, immediate execution-only tasks. See "Don't Use When" section for details. Outputs: Structured XML/Markdown prompt, quality score (before/after), optional team brief + per-agent sub-prompts, agent team output files. Success criteria: Single mode quality score ≥ 7/10; Repromptception per-agent prompt quality score 8+/10; all required sections present, actionable and specific.
adaptive-compaction
IncludedAdaptive add-on policy and recovery layer that decides WHEN to compact, prune, snapshot, or fork -- replacing fixed-percent auto-compaction across Claude Code, Codex, and MCP-capable hosts. Trigger on auto-compact timing or damage: "when should I compact", "is it safe to compact now or start a fresh session", "auto-compact fires too early/mid-task", "switching to an unrelated task but the window still has space", "context rot", "answers get worse the longer the session runs", "the agent forgot the plan or my decisions after it summarized", "add a layer on top that manages context without changing the agent", raising autoCompactWindow to give the policy room, or installing/tuning a cross-tool compaction policy or PreCompact hook -- even when "compaction" is never said but the problem is context-window pressure or post-summarization memory loss. Do NOT use to summarize a conversation, build RAG, write a summarization prompt (decides WHEN not HOW), or answer max-context-length trivia.
agent-skill-creator
IncludedCreate cross-platform agent skills from workflow descriptions. Activates when users ask to create an agent, automate a repetitive workflow, create a custom skill, or need advanced agent creation. Triggers on phrases like create agent for, automate workflow, create skill for, every day I have to, daily I need to, turn process into agent, need to automate, create a cross-platform skill, validate this skill, export this skill, migrate this skill. Supports single skills, multi-agent suites, transcript processing, template-based creation, interactive configuration, cross-platform export, and spec validation.
llm-wiki
IncludedUse when building or maintaining a persistent personal knowledge base (second brain) in Obsidian where an LLM incrementally ingests sources, updates entity/concept pages, maintains cross-references, and keeps a synthesis current. Triggers include "second brain", "Obsidian wiki", "personal knowledge management", "ingest this paper/article/book", "build a research wiki", "compound knowledge", "Memex", or whenever the user wants knowledge to accumulate across sessions instead of being re-derived by RAG on every query.
skill-master
IncludedAgent Skills authoring, evaluation, and optimization. Create, edit, validate, benchmark, and improve skills following the agentskills.io specification. Use when designing SKILL.md files, structuring skill folders (references, scripts, assets), ingesting external documentation into skills, running trigger evals, benchmarking skill quality, optimizing descriptions, or performing blind A/B comparisons. Keywords: agentskills.io, SKILL.md, skill authoring, eval, benchmark, trigger optimization.