rewrite-rubric
How to rewrite a selected passage per the user's instruction without silently expanding scope, breaking length, or losing quoted material
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
菜单
How to rewrite a selected passage per the user's instruction without silently expanding scope, breaking length, or losing quoted material
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
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 | rewrite-rubric |
| description | How to rewrite a selected passage per the user's instruction without silently expanding scope, breaking length, or losing quoted material |
| when_to_use | The user asks to "rewrite this", "fix this", "tighten this", "make this clearer", "make this more formal", "fix the typo", or otherwise applies an edit to the currently-selected text. Also fires when the user pastes a passage with an edit instruction. |
Rewriting is edit work, not authoring. The user has chosen the words they want changed; your job is to honor the change they asked for and nothing more.
Before producing the rewrite, always:
recall_writing_style — load the voice rulebook. Voice is preserved even on a minimal edit.get_current_selection — read the actual selected text. Never rewrite from memory or paraphrase the user's selection.rewrite_passage_hints(text, instruction) — get objective metrics and the classified_ask. The classified_ask is the contract for this turn.If get_current_selection returns {"status": "no_selection"}, ask the user to highlight the passage first — never rewrite the whole document on a vague "fix this".
The classified_ask from rewrite_passage_hints tells you the shape of the edit:
| classified_ask | Allowed | NOT allowed |
|---|---|---|
minimal-fix | Change only the broken token (typo, grammar slip). | Touching anything that wasn't wrong. |
shorten | Cut up to ~30-45% of words. | Adding new ideas. |
expand | Add detail or examples up to ~60% more. | Padding with hedges or filler. |
clarify | Restructure or simplify sentences. | Changing the argument or claim. |
raise-register | More formal word choice, fewer contractions. | Adding pomp the user wouldn't use. |
lower-register | Looser, contractions, friendlier tone. | Slang the user has never used in style memory. |
rewrite-preserve-meaning | Voice-match rewrite at same length. | Drifting on meaning to chase prettiness. |
unspecified / other | Ask the user what they want. | Guessing. |
The hints tool returns target_word_count_low and target_word_count_high. The rewrite must land inside that window. If it doesn't, trim or expand before returning.
rewrite_passage_hints returns quoted_examples and a count of citation_markers. These are inviolable:
If the surrounding document uses specific terminology (get_full_document reveals it), the rewrite must keep it:
The user's terminology is part of their voice.
[TK: figure]), surface that separately in the chat reply after the passage block.minimal-fix; honor it.clarify or raise-register. Stay within the window.