grow-canvas

Propose new Mica canvas cards from conversation signals.

Updated May 21, 2026
One-click install
npx skills add https://github.com/mica-agent/mica-solar-system --skill grow-canvas
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: grow-canvas
Source: https://github.com/mica-agent/mica-solar-system/tree/main/.qwen/skills/grow-canvas
Command: npx skills add https://github.com/mica-agent/mica-solar-system --skill grow-canvas

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It prevents your Mica canvas from becoming either empty (missing key context) or cluttered (filled with placeholder docs) by proposing the right new card only when the conversation reveals a concrete need.

Core Features & Use Cases

  • On-demand card proposals: Suggests new surface areas like todo lists, decision logs, flows, architecture diagrams, data models, READMEs, or terminal access when the right signal appears.
  • Signal-driven creation rules: Uses the first matching trigger, proposes only one card per turn, and avoids duplicating existing cards.
  • Context-aware seeding: Seeds newly created cards with the actual details already discussed, and links related sibling docs when overlaps exist.

Quick Start

Ask the AI to propose a new canvas card when your discussion starts accumulating actionable steps, tradeoffs, user flows, system boundaries, or data models.

Frequently Asked Questions about grow-canvas

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I create canvas cards on-demand during product planning discussions?

You can automatically generate canvas cards by discussing planning, decisions, user flows, or architecture; the Skill detects these signals and proposes one relevant card per turn seeded with your conversation context.

What is the best way to prevent my Mica canvas from becoming cluttered with placeholder documentation?

To avoid canvas clutter, use signal-driven creation rules that propose new cards only when a concrete need emerges in the conversation, preventing duplicate or upfront placeholder documents.

How does context-aware seeding work when proposing new architecture diagrams and data models?

Context-aware seeding populates newly proposed architecture diagrams and data models by extracting actual details from your ongoing discussion, automatically linking related sibling docs when overlaps exist.

Does this Skill require an existing mermaid rendering class to propose canvas cards?

Yes, this Skill requires an existing built-in mermaid rendering class via .mmd files to ensure proposed architecture diagrams and user flows render correctly on the Mica canvas.

Can I scaffold multiple todo lists and decision logs upfront in a single turn?

No, you cannot scaffold multiple cards upfront; the Skill enforces strict one-proposal-per-turn behavior, applying the first matching trigger to incrementally propose todo lists or decision logs as needs emerge.