feat(brainstorming): Smart question batching, visual previews, alternative surfacing#520
Open
feat(brainstorming): Smart question batching, visual previews, alternative surfacing#520
Conversation
…ative surfacing Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
|
No actionable comments were generated in the recent review. 🎉 📝 WalkthroughWalkthroughReplaces a one-question-at-a-time clarifying workflow with a smart-batching approach: classify questions as text or visual, ask text questions first in batches (≤4 independent text questions per AskUserQuestion call, multiple batches allowed), then ask visual questions one-at-a-time with ASCII mockups. Changes
Sequence Diagram(s)sequenceDiagram
participant Agent as Brainstorming Skill
participant User as User
participant API as AskUserQuestion API
participant Renderer as ASCII Preview Renderer
Agent->>Agent: Classify clarifying questions\n(text vs visual)
Agent->>API: Send batch of up to 4 text questions
API->>User: Present batched text questions
User-->>API: Respond to batch
API-->>Agent: Return responses
Agent->>Agent: If more text questions -> repeat batch
Agent->>Renderer: Prepare ASCII mockup for next visual question
Renderer->>User: Show ASCII preview (one visual question)
User-->>Agent: Respond to visual question
Agent->>Agent: Continue visual questions one-at-a-time until done
Estimated code review effort🎯 2 (Simple) | ⏱️ ~8 minutes Poem
🚥 Pre-merge checks | ✅ 3✅ Passed checks (3 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 3
🧹 Nitpick comments (1)
skills/brainstorming/SKILL.md (1)
69-69: Optional: replace ALL CAPS with bold for consistent emphasis styleEvery other emphasis in this document uses
**bold**; ALL CAPS is an outlier.✏️ Proposed fix
- - Ask visual questions ONE AT A TIME, using the `markdown` preview field on each option for ASCII mockups + - Ask visual questions **one at a time**, using the `markdown` preview field on each option for ASCII mockups🤖 Prompt for AI Agents
Verify each finding against the current code and only fix it if needed. In `@skills/brainstorming/SKILL.md` at line 69, Replace the ALL CAPS emphasis in the line "Ask visual questions ONE AT A TIME, using the `markdown` preview field on each option for ASCII mockups" with the document's standard bold style; edit that string in SKILL.md and change "ONE AT A TIME" to a bolded phrase (e.g., **one at a time**) so emphasis uses **bold** instead of ALL CAPS for consistency.
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In `@skills/brainstorming/SKILL.md`:
- Line 62: Replace the phrase "multiple choice questions" with the hyphenated
compound adjective "multiple-choice questions" in SKILL.md (look for the exact
string "Prefer multiple choice questions when possible, but open-ended is fine
too") so the compound modifier preceding the noun is grammatically correct.
- Line 100: The bold phrase "**Multiple choice preferred**" in SKILL.md should
use the same hyphenation as the earlier instance — change it to
"**Multiple-choice preferred**" to be consistent; update the bold text in that
list item and scan for other occurrences of "Multiple choice" to normalize to
"Multiple-choice" where used in headings or list items (look for the bolded list
entry and other list entries around lines with the multiple-choice guidance).
- Line 97: The summary line "Batch independent questions — up to 4 per
AskUserQuestion call, multiple batches if needed" is ambiguous; update it to
match the concrete threshold used earlier ("if more than 4") by changing the
trailing phrase to something explicit like "multiple batches if more than 4
questions need asking" (referencing the "Batch independent questions" heading
and the earlier trigger on line 67) so the principle clearly specifies when
batching occurs.
---
Nitpick comments:
In `@skills/brainstorming/SKILL.md`:
- Line 69: Replace the ALL CAPS emphasis in the line "Ask visual questions ONE
AT A TIME, using the `markdown` preview field on each option for ASCII mockups"
with the document's standard bold style; edit that string in SKILL.md and change
"ONE AT A TIME" to a bolded phrase (e.g., **one at a time**) so emphasis uses
**bold** instead of ALL CAPS for consistency.
- multiple choice → multiple-choice (compound adjective hyphenation) - ONE AT A TIME → **one at a time** (bold instead of ALL CAPS) - "if needed" → "if more than 4" (match concrete threshold) Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
There was a problem hiding this comment.
Actionable comments posted: 1
🤖 Prompt for all review comments with AI agents
Verify each finding against the current code and only fix it if needed.
Inline comments:
In `@skills/brainstorming/SKILL.md`:
- Around line 97-98: Update the Key Principles wording to make it explicit that
the 4-per-AskUserQuestion limit applies only to text questions: change "Batch
independent questions — up to 4 per AskUserQuestion call, multiple batches if
more than 4" to explicitly say "Batch independent text questions — up to 4 per
AskUserQuestion call, multiple batches if more than 4", keeping the existing
"Visual questions last, one at a time — use `markdown` preview field for ASCII
mockups" line intact so the distinction between AskUserQuestion text batching
and single visual questions is unambiguous.
"Batch independent questions" → "Batch independent text questions" to avoid ambiguity with the one-at-a-time visual question rule. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
obra
added
enhancement
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Improve the brainstorming skill's clarifying questions phase with smart batching, visual previews, and alternative surfacing.
Motivation and Context
When brainstorming a feature with 6+ questions, the current "one question at a time" approach creates excessive round-trips. Additionally, layout/UI questions use text-only descriptions ("Full-width stacked" vs "Two-column") when the AskUserQuestion tool already supports
markdownpreviews for ASCII mockups. Finally, questions with ambiguous concepts (e.g., "reference ID" vs "payment method") don't always surface the alternatives as choosable options.How Has This Been Tested?
Tested on a Guardian Financial Overview scenario (6 clarifying questions). 3 runs per version (6 total, Sonnet).
Three invariants held across all improved runs:
markdownpreview mockups (never text-only)The only variance between improved runs was which text questions got batched together — the pattern itself was 100% consistent.
Breaking Changes
None. The skill still produces AskUserQuestion calls with the same schema. Users see fewer interaction rounds and richer visual options.
Types of changes
Checklist
Additional context
The diff is minimal — replaces the "one question at a time" bullet points with a "smart batching" section (6 bullet points) and updates Key Principles to match. No structural changes to the rest of the skill.
Example: Layout question output — before vs after
Before (text-only):
{ "label": "Full-width stacked", "description": "All sections stack vertically in a single column" }After (ASCII mockup in markdown preview):
{ "label": "Full-width stacked layout", "description": "All sections stack vertically in a single column.", "markdown": "┌────────────────────────────────────────────┐\n│ ┌──────────┐ ┌──────────┐ ┌──────────┐ │\n│ │ Balance │ │ Charges │ │ Payments │ │\n│ └──────────┘ └──────────┘ └──────────┘ │\n│ > Child A — charges by group │\n│ Transaction history [Filter v] │\n│ [Record Payment] [Record Adjustment] │\n└────────────────────────────────────────────┘" }Summary by CodeRabbit