Menu
How to draft email replies and new messages that sound like the user, using the style memory
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
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.
Pattern for turning a brief into a coherent deck outline — intent detection, section ratios, draft headings, expansion
| name | voice-matching |
| description | How to draft email replies and new messages that sound like the user, using the style memory |
| when_to_use | The user asks to draft, write, reply to, compose, or "knock out" an email. Includes "make me a reply to this" and "send Anna a note about X". |
The goal is a draft the user could send unchanged. Not "a good email" — their good email.
Every drafting call starts with:
recall_writing_style — load the voice rulebook. If empty, draft conservatively (short, plain, no flourishes) and tell the user "I haven't learned your style yet — this is generic; reply with edits and I'll learn."get_current_compose if the user is replying or building on something already in the compose window. Don't blow away their existing text — extend it.get_selected_message if the request is "reply to this".The style memory typically contains rules like:
Hi <name>, vs Hey <name> vs none).Thanks, vs Best, vs — J).Apply them mechanically. If the memory says "no exclamation marks" and the draft has one, fix it before returning.
recall_memory(topic="recipient-<email>") before drafting to someone the user emails often.When the agent inserts a draft into the compose window:
Subject: line in the body. Subject goes in the subject field separately.When the agent returns a draft for the user to see in the chat first (no compose open):
After the user confirms a send (the taskpane fires learn_style_from_sent_email), the agent treats it as a teaching signal — especially if the user edited the draft before sending. The edit is the gold-standard voice sample.