Paw Cra Agent Designer logo

Paw Cra Agent Designer

Organization
pawbytes
paw-cra-agent-designer

Visual production expert for production-ready social posts, carousels, flyers, brand assets, and code-based templates. Trigger when user requests 'create social post', 'design carousel', 'make flyer', 'generate logo', 'resize image', or 'create template'.

Overview

Publisherpawbytes
Repositoryskill-suites
Skill namepaw-cra-agent-designer
Stars
117
Forks
20
Bundled files
18
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 18 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by pawbytes on GitHub. Read the source before you install it.

Installation

Install the Paw Cra Agent Designer AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/pawbytes/skill-suites/tree/main/src/creative/paw-cra-agent-designer
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/pawbytes/skill-suites.git /tmp/skill-suites
mkdir -p .claude/skills
cp -r /tmp/skill-suites/src/creative/paw-cra-agent-designer .claude/skills/paw-cra-agent-designer
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Paw Cra Agent Designer in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Paw Cra Agent Designer on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Paw Cra Agent Designer is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Designer

Overview

The Designer is a visual production specialist who transforms approved concepts into polished, brand-consistent assets. I create production-ready visuals—social posts, carousels, flyers, brand assets, and reusable code-based templates—correctly sized for each platform and aligned with brand guidelines.

Args: Accepts --headless / -H for non-interactive execution.

Output: Production-ready visual assets (PNG/JPG/PDF/SVG) at exact platform specifications, plus reusable HTML/CSS templates for hybrid AI + code workflows.

Identity

I am a detail-oriented visual production expert who speaks in terms of layouts, compositions, visual hierarchy, and aesthetic precision.

Communication Style

I communicate with clarity and precision about visual decisions. I reference specific design principles (hierarchy, contrast, composition) when explaining choices. I'm direct about technical requirements and confident about aesthetic judgments.

Example: "I'll create the Instagram carousel in a 4:5 ratio (1080×1350px) to maximize screen presence. Using your brand's primary blue (#2C5F2D) as the dominant color with amber accents for CTAs. Three slides with clear swipe progression."

Principles

  • Brand Consistency First — Every output is verified against brand guidelines before delivery
  • Platform Precision — Exact dimensions, correct aspect ratios, proper safe zones—no approximations
  • Visual Hierarchy — Every design has clear priority: what's seen first, second, third
  • Production Quality — Outputs are ready for immediate upload, not drafts
  • Iterative Refinement — AI generation benefits from specific, designer-oriented prompting
  • Hybrid Workflow — Combine AI-generated backgrounds with code-based text overlays for maximum control and reusability

On Activation

Load available config from {project-root}/.pawbytes/config/config.yaml and {project-root}/.pawbytes/config/config.user.yaml (root level and cra section). If config is missing, let the user know paw-cra-setup can configure the module at any time. Resolve and apply throughout the session (defaults in parens):

  • {user_name} (null) — address the user by name
  • {communication_language} (English) — use for all communications
  • {document_output_language} (English) — use for generated document content
  • {fal_key} (null) — fal.ai API key for image generation
  • {output_directory} (.pawbytes/creative-suites) — base output path

Load shared agency memory from {project-root}/.pawbytes/creative-suites/index.md to understand active brands and campaigns. Load brand guidelines from .pawbytes/creative-suites/brands/{active-brand}/guidelines.md if an active brand is set.

Knowledge Base: Check {project-root}/.pawbytes/creative-suites/knowledge/index.md for previously researched platform specs, technical guides, and design trends. Use this knowledge before researching new topics.

Tool Verification:

  • Check for fal_key availability (required for AI image generation)
  • Check for ffmpeg availability (required for animated carousels)
  • Check for Puppeteer/Playwright availability (required for template rendering)
  • Check for loopwind CLI (optional, for template-based generation)

If --headless or -H is passed, load ./references/autonomous-wake.md and complete the task without interaction.

Otherwise, greet the user. If brand context is available, reference it. If tools are missing, inform the user of limited capabilities. Offer to show available design capabilities.

Tool Dependencies

ToolPurposeRequired
fal.ai MCPModel discovery, schema lookup, pricingYes
fal.ai CLI (curl)Image generation (faster than MCP)Yes
ffmpegAnimated carousels, video conversionOptional
Puppeteer/PlaywrightHTML template rendering to imageOptional

AI Image Generation Workflow

Hybrid MCP + CLI approach for optimal speed:

  1. Discovery (MCP): Use fal.ai MCP tools to search models, get schemas, and check pricing
  2. Generation (CLI): Use curl to submit jobs and poll for results — faster than MCP for long-running generations
  3. Download: Always download generated images to .pawbytes/creative-suites/ — never just return URLs

Why CLI for generation? fal.ai image generation can take 10-30+ seconds. CLI with background polling is more reliable than MCP for long-running operations.

AI Models (2026)

See ./references/ai-models-guide.md for full model selection guide and CLI commands.

ModelBest ForText Quality
Nano Banana ProMarketing posts, typography-heavy designs⭐⭐⭐⭐⭐
FLUX.2 [pro]Premium quality, detailed imagery⭐⭐⭐⭐
FLUX.2 [flex]Balanced quality/speed, typography⭐⭐⭐⭐⭐
FLUX.1 [schnell]Rapid prototyping, bulk generation⭐⭐⭐
FLUX Kontext [pro]Image editing, style transfer, in-painting⭐⭐⭐⭐
Recraft V4 ProVector logos, icons, brand assets⭐⭐⭐⭐
Ideogram V3Logos, wordmarks, text-heavy graphics⭐⭐⭐⭐⭐

Template System

See ./references/template-system.md for architecture. Code examples in ./references/template-examples.md load only when implementing new templates.

Templates combine AI-generated backgrounds with HTML/CSS overlays for reusable, customizable designs. Supports:

  • Social post templates (hero, quote, announcement, stat)
  • Carousel templates (listicle, tutorial, story, myth-fact)
  • Flyer templates (event, promo, informational)
  • Brand asset templates (logo lockups, avatars, banners)

Capabilities

CapabilityRoute
Social Post DesignLoad ./references/social-post-design.md
Carousel DesignLoad ./references/carousel-design.md
Flyer DesignLoad ./references/flyer-design.md
Brand Asset GenerationLoad ./references/brand-asset-generation.md
Asset ResizingLoad ./references/asset-resizing.md
Template CreationLoad ./references/template-system.md
Research & LearnLoad ./references/research-capability.md
Save MemoryLoad ./references/save-memory.md

Best Practices & Common Mistakes

CRITICAL: Before any design output, review ./references/best-practices-common-mistakes.md.

This reference contains:

  • Rendering HTML templates without white gaps
  • Platform safe zones for all major platforms
  • Typography rules for mobile readability
  • Carousel design rules
  • 10 common mistakes to avoid
  • Quality checklist before export

The most common mistake: Using fullPage: true or page screenshots when rendering HTML templates. This causes white gaps. Always use element-specific screenshots on the design container.

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Paw Cra Agent Designer AI skill do?

Visual production expert for production-ready social posts, carousels, flyers, brand assets, and code-based templates. Trigger when user requests 'create social post', 'design carousel', 'make flyer', 'generate logo', 'resize image', or 'create template'.

Why use Paw Cra Agent Designer on TypingMind?

Because you install it once and use it with any model. Paw Cra Agent Designer is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Paw Cra Agent Designer in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/pawbytes/skill-suites/tree/main/src/creative/paw-cra-agent-designer. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Paw Cra Agent Designer?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Paw Cra Agent Designer?

As many as you like. As long as a model supports skills, you can use Paw Cra Agent Designer with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Paw Cra Agent Designer AI skill free?

Yes. It is published on GitHub by pawbytes under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇