[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill-34c29138-7c29-4e2c-a795-a7c295e75489":3,"$fTPKZzmbdg-mde9Scz6WYTwpNSdj7rCyZDjVYkeY4BJU":43},{"id":4,"title":5,"description":6,"categoryId":7,"moduleId":8,"tags":9,"prompt":10,"icon":11,"source":12,"sourceUrl":13,"authorId":14,"authorName":15,"isPublic":16,"stars":17,"runs":18,"createdAt":19,"updatedAt":19,"module":20,"category":27,"packages":34},"34c29138-7c29-4e2c-a795-a7c295e75489","rich-elicitation","在开始模糊任务之前，通过多轮提问进行澄清。当两个以上的任务维度各有三个以上的可行答案时，就采取行动。","cat_writing_academic","mod_writing","sickn33,writing","---\nname: rich-elicitation\ndescription: \"Asks clarifying questions in multiple rounds before starting ambiguous tasks. Fires when 2+ task dimensions each have 3+ viable answers.\"\ncategory: productivity\nrisk: none\nsource: self\nsource_type: self\ndate_added: \"2026-05-07\"\nauthor: abubakar\ntags: [elicitation, clarifying-questions, ambiguity, multi-round, prompt-engineering]\ntools: [antigravity]\n---\n\n# Rich Elicitation Skill\n\n## Overview\n\nThis skill governs how Antigravity resolves task ambiguity before starting work. When a user's request has too many unanswered dimensions — each with several reasonable answers — Antigravity asks targeted clarifying questions across multiple rounds rather than silently picking defaults.\n\nThe goal is a correct first draft, not a generic answer that requires three revision cycles. Rounds are capped at three; anything still unclear after Round 3 gets a stated assumption and Antigravity proceeds.\n\n---\n\n## When to Use This Skill\n\n- Use when a request has 2 or more dimensions that are ambiguous and each has 3+ viable options\n- Use when the user's likely intent is unclear across scope, audience, tone, format, or strategy\n- Use when an early answer would meaningfully change the structure or direction of the output\n- Use when working on writing, planning, design, recommendations, or creative tasks with open-ended scope\n- Use when a Round 1 answer unlocks a new set of meaningful choices that need resolving before proceeding\n\nDo **not** trigger for:\n- Simple factual lookups or math\n- Clearly scoped requests with a single obvious interpretation\n- Minor unknowns where a safe default exists\n\n---\n\n## How It Works\n\n### Step 1: Run the Trigger Checklist\n\nBefore starting any task, mentally check how many of these apply:\n\n| Signal | Action |\n|---|---|\n| Multiple valid output formats | Ask about format |\n| Audience is unknown | Ask about audience |\n| Tone is ambiguous | Ask about tone |\n| Scope could be narrow or broad | Ask about depth\u002Flength |\n| Technical vs. simple treatment unclear | Ask about technical level |\n| Multiple strategic directions exist | Ask which direction |\n| User's constraints are unknown | Ask about constraints |\n\n**If 2+ rows apply → trigger this skill.**\n\n### Step 2: Ask Round 1 Questions\n\nAsk up to 3 questions using `ask_user_input_v0`. Group related questions in a single call. Lead with 1–2 sentences explaining why you're asking. Mark one option per question as **(Recommended)**.\n\n### Step 3: Re-run the Checklist\n\nAfter Round 1 answers, re-run the checklist on what's still unresolved. If 2+ rows still apply, run Round 2. Otherwise, proceed.\n\n### Step 4: Run Follow-up Rounds (if needed)\n\n| Round | Purpose | Max questions |\n|---|---|---|\n| Round 1 | Blocking questions — shape the entire output | 3 |\n| Round 2 | Follow-ups unlocked by Round 1 answers | 3 |\n| Round 3 | Final details — use sparingly | 2 |\n\nTransition between rounds naturally. Don't announce \"Round 2\" mechanically. Use phrasing like:\n> \"Got it — that helps a lot. One more thing before I start:\"\n\n### Step 5: Proceed\n\nAfter Round 3 (or earlier, if enough context exists), state any remaining assumptions briefly and begin the task.\n\n---\n\n## Examples\n\n### Example 1: Multi-Round — Investor Pitch Deck\n\n**User:** \"Help me create a presentation for my project.\"\n\n**Round 1** — Three blocking questions:\n\n```\nFraming: \"This could go a lot of different ways. Quick questions before I start:\"\n\nQ1: Who is the audience?\n  - Internal team \u002F colleagues\n  - External clients or partners\n  - Investors or stakeholders (Recommended)\n  - General \u002F public audience\n\nQ2: What's the primary goal?\n  - Inform and update\n  - Persuade and drive a decision (Recommended)\n  - Teach or explain a concept\n  - Pitch and raise funding\n\nQ3: How much content do you already have?\n  - Starting from scratch\n  - Rough notes or an outline (Recommended)\n  - Full draft — just needs polish\n  - Previous version to update\n```\n\n*User answers: Investors | Pitch | Rough notes → Round 2 is warranted.*\n\n**Round 2** — Unlocked by Round 1:\n\n```\nFraming: \"Perfect — investor pitch it is. A couple more things:\"\n\nQ1: What stage is this raise?\n  - Pre-seed \u002F idea stage\n  - Seed round (Recommended)\n  - Series A or later\n  - Strategic partnership \u002F grant\n\nQ2: How long should the deck be?\n  - Short and punchy, 8–10 slides (Recommended)\n  - Standard, 12–15 slides\n  - Comprehensive, 20+ slides\n```\n\n*No Round 3 needed — proceed.*\n\n---\n\n### Example 2: Single-Round — Business Email\n\n**User:** \"Write a business email to follow up on a proposal.\"\n\n**Round 1 only:**\n\n```\nFraming: \"Two quick questions to nail the tone:\"\n\nQ1: What tone should this email strike?\n  - Formal and professional (Recommended)\n  - Friendly but direct\n  - Urgent and firm\n  - Warm and relationship-focused\n\nQ2: What's the primary goal?\n  - Request action \u002F get a response (Recommended)\n  - Share information only\n  - Repair or maintain the relationship\n  - Negotiate or push back\n```\n\n*Enough context. No Round 2 needed.*\n\n---\n\n## Best Practices\n\n- ✅ Always mark one option per question as **(Recommended)**\n- ✅ Lead with a 1–2 sentence framing before the question widget\n- ✅ Group up to 3 related questions in a single `ask_user_input_v0` call\n- ✅ Re-evaluate after each round — stop as soon as you have enough context\n- ✅ Use `single_select` for mutually exclusive choices, `multi_select` when combinations are valid\n- ✅ State remaining assumptions explicitly before proceeding after Round 3\n- ❌ Don't ask 6 separate question calls when 2 grouped calls would do\n- ❌ Don't mark two options as Recommended in the same question\n- ❌ Don't use vague option labels like \"Other\" or \"It depends\" without elaborating\n- ❌ Don't mechanically label rounds in the UI (\"Round 1:\", \"Round 2:\")\n- ❌ Don't run a follow-up round for minor details that have safe defaults\n\n---\n\n## Limitations\n\n- This skill does not validate whether the user's answers are internally consistent — it trusts them as given.\n- Round structure is a guideline, not a rigid contract; judgment is required on when to stop.\n- Works best with `ask_user_input_v0` — in environments without that tool, question quality may degrade.\n- Does not handle tasks where ambiguity can only be resolved by fetching external information (e.g., reading a file the user hasn't uploaded).\n- Not designed for real-time or high-latency-sensitive workflows where any question overhead is unacceptable.\n\n---\n\n## Security & Safety Notes\n\nThis skill is pure reasoning — it issues no shell commands, reads no files, makes no network requests, and mutates no state. Risk level is `none`.\n\nNo `npm run security:docs` review is required for this skill.\n\n---\n\n## Common Pitfalls\n\n- **Problem:** Antigravity asks one good question, gets an answer, then proceeds without checking if new unknowns emerged.\n  **Solution:** Always re-run the trigger checklist mentally after each round before deciding to proceed.\n\n- **Problem:** All options in a question look equally valid so Antigravity marks none as Recommended.\n  **Solution:** Pick the option that works for most users or is lowest-risk and mark it. \"No preference\" is rarely true.\n\n- **Problem:** Antigravity runs 4+ rounds trying to eliminate every unknown.\n  **Solution:** Hard cap at 3 rounds. After Round 3, state assumptions and proceed.\n\n- **Problem:** Round 2 questions cover the same category as Round 1 (e.g., tone again).\n  **Solution:** Each round should unlock new dimensions, not re-ask resolved ones.\n\n---\n\n## Related Skills\n\n- `@ask-user-questions` — Single-round elicitation with recommended options. Use that skill for simpler tasks; use rich-elicitation when answers to early questions open up new meaningful choices.\n","","imported","https:\u002F\u002Fgithub.com\u002Fsickn33\u002Fantigravity-awesome-skills","user_system_seed","SkillOPIC",true,170,998,"2026-05-16 13:37:14",{"id":8,"name":21,"slug":22,"icon":23,"description":24,"sort":25,"createdAt":26},"写作研究","writing","mdi-pencil-outline","从学术写作到创意文案，让 AI 成为你的专属写作助手",1,"2026-05-16 12:53:40",{"id":7,"name":28,"slug":29,"icon":30,"description":31,"moduleId":8,"sort":32,"skillCount":33,"createdAt":26},"学术研究","academic","mdi-book-open-variant","论文润色、文献综述、摘要提取",2,8,[35],{"id":36,"skillId":4,"version":37,"fileName":38,"fileSize":39,"filePath":40,"fileHash":41,"manifest":42,"createdAt":19},"127ba71c-15af-4acb-a503-819068492477","1.0.0","rich-elicitation.zip",3512,"uploads\u002Fskills\u002F34c29138-7c29-4e2c-a795-a7c295e75489\u002Frich-elicitation.zip","3d4869cdbc0dbcae03382a72618848c22d1fc1683946b8ef33f75befaf9b3c7a","[{\"path\":\"SKILL.md\",\"isDirectory\":false,\"size\":7766}]",{"code":44,"message":45,"data":46},200,"success",{"items":47,"stats":48,"page":51},[],{"averageRating":49,"totalRatings":49,"ratingCounts":50},0,[49,49,49,49,49],{"limit":52,"offset":49,"hasMore":53,"nextOffset":52,"ratedOnly":16},15,false]