home/business/prompt-strategy

Prompt Strategy

GPTClaudeGemini··816 copies·updated 2026-07-14
prompt-strategy.prompt
# Frame Prompt Strategy

Purpose: load this when Frame needs tool-specific prompt wording, chaining order, or scope-control patterns for Figma MCP calls.

## Contents

- [General Principles](#general-principles)
- [get_design_context](#get_design_context)
- [get_variable_defs](#get_variable_defs)
- [get_screenshot](#get_screenshot)
- [get_metadata](#get_metadata)
- [whoami](#whoami)
- [get_code_connect_map](#get_code_connect_map)
- [get_code_connect_suggestions](#get_code_connect_suggestions)
- [add_code_connect_map / send_code_connect_mappings](#add_code_connect_map--send_code_connect_mappings)
- [create_design_system_rules](#create_design_system_rules)
- [get_figjam](#get_figjam)
- [generate_diagram](#generate_diagram)
- [Tool Chaining Patterns](#tool-chaining-patterns)

## General Principles

1. **Be specific about what you need** — Vague prompts return verbose, unfocused data
2. **Specify depth when possible** — Avoid deep recursive extraction when shallow data suffices
3. **Name the output format** — Tell the tool how you want data structured
4. **One concern per call** — Don't ask for everything in one request; layer calls by priority

---

## get_design_context

### Purpose
Extract component structure, styles, Auto Layout properties, and hierarchy.

### Effective Patterns

**Component Analysis (Focused)**

when to use it

Community prompt sourced from the open-source GitHub repo simota/agent-skills (MIT). A "Prompt Strategy" style prompt — adapt the placeholders and specifics to your task. Imported as-is and not independently retested here, so check the output before relying on it.

tags

businesscommunitygeneral

source

simota/agent-skills · MIT