deck-outlining
Pattern for turning a brief into a coherent deck outline — intent detection, section ratios, draft headings, expansion
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
メニュー
Pattern for turning a brief into a coherent deck outline — intent detection, section ratios, draft headings, expansion
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
Use when the user wants to build a Google ADK agent — scaffolding a new agent, adding tools/skills/prompts to one, or asking how to structure an ADK project. Triggers on phrases like "create an ADK agent", "build a Google agent", "scaffold an agent", "make an agent that does X", "I need an ADK skeleton", or any task involving generated-agents/, LlmAgent, SkillToolset, or the ADK framework. Also use whenever the user mentions `nuvel`, the `nuvel` CLI, or asks about agent architecture patterns / callbacks / HITL / streaming / ADK prompt engineering — nuvel ships the canonical knowledge skills for those topics. Lean toward triggering — if the task touches Google ADK at all, this skill is in scope.
Delegate work to sub-agents with the ADK 2.0 Task API — `mode='task'`, `mode='single_turn'`, `mode='chat'` on `LlmAgent`, the auto-attached `finish_task` tool, and typed contracts via `input_schema` / `output_schema`. Load this skill when one agent needs to hand a bounded unit of work to another and get a validated result back, or when migrating off SequentialAgent / ParallelAgent / LoopAgent.
Creating valid SKILL.md files following the agentskills.io specification — frontmatter, instructions, references directory, progressive disclosure (L1/L2/L3), and SkillToolset wiring in agent.py. Load this skill when generating domain skills for an agent.
Build graph-based agents with ADK 2.0 `Workflow` — declare nodes and edges, route conditionally, fan-out/fan-in in parallel, run dynamic nodes at runtime, and add human-in-the-loop revision cycles. Load this skill whenever the agent needs anything beyond a strictly linear or trivially parallel pipeline.
Agent architecture patterns for Google ADK 2.0 — when to reach for a single LlmAgent, a Workflow graph (new default for multi-step orchestration), or the shortcut classes SequentialAgent / LoopAgent / ParallelAgent. Load this skill when deciding the agent's top-level shape.
When and how to suggest reordering, splitting, or trimming a deck — agenda placement, CTA placement, problem-solution arc, methodology-results arc
| name | deck-outlining |
| description | Pattern for turning a brief into a coherent deck outline — intent detection, section ratios, draft headings, expansion |
| when_to_use | The user asks to outline, plan, scaffold, or draft a new deck from a brief. Includes "make me a deck about X", "outline a pitch for Y", "what slides do I need for Z". |
The goal is an outline the user can walk through and say "yes, that arc". Not a generic skeleton; a deck shaped by what kind of deck this is.
Every outlining call starts with:
recall_deck_style — load the user's deck-style rulebook. If empty, use defaults (3-5 bullets per slide, <=10 words each, parallel verb-led phrasing) and tell the user "I haven't learned your deck style yet — this is generic; keep what works and I'll learn."plan_deck_outline(brief, target_slide_count=10) — get the intent (pitch / training / report / status / general), the intro/body/closing slide budget, and intent-specific hints.If the user gave no slide count, default to 10. If the brief mentions a constraint ("3-minute pitch" → ~5 slides, "30-minute training" → ~15), respect it.
plan_deck_outline does this from keywords. If the result feels wrong (the brief mentions "pitch" but the user clearly means a status update), override the intent in your draft, but say so explicitly: "I read this as a status update, not a pitch — tell me if you want pitch shaping instead."{"intro": N, "body": N, "closing": N}. Build that many slides in each section; don't pad and don't trim without saying why.Return the outline as a structured block:
Slide 1 — Title: <title>
Bullets:
- <bullet>
- <bullet>
Notes: <2-4 lines>
Slide 2 — Title: …
…
The taskpane parses this to surface per-slide "Insert as new slide" buttons. Keep titles, bullets, and notes on their own lines.
If the brief is one sentence ("a deck about onboarding"), ask exactly one clarifying question — audience or duration — and proceed with reasonable defaults if the user doesn't answer. Don't outline blind on a vague brief.