Art & Design

Agent guidance files (AGENTS.md, WARP.md) for awesome-claude-skills

Design loops, don't prompt agents — a Claude Code skill that writes a self-verifying LOOP.md grounded in your real repo & agent.

A community Codex skill pack that guides agents to design architecture, file boundaries, patterns, and tests before implementation.

A Codex/Claude skill for deciding when and how to craft HTML artifacts instead of Markdown.

Reusable Codex and Claude skill for launching B2C mobile app businesses end to end

Codex/Claude skill for designing robust scraping, API, Playwright, and authorized data acquisition pipelines

Long-term pregnancy health assistant built on LLM + Agent workflows

Requirements Engineering corpus for AI — Claude Code skill, plugin marketplace, and MCP server (Claude Desktop, Cursor, Cline, Continue, Zed, OpenAI Responses). Methodology from IFPB ERS course + canonical bibliography (Sommerville, Pressman, Wiegers, Cohn, BABOK v3, SBC 002/2024). Bilingual: en-CA + pt-BR snapshot. CC BY-SA 4.0.

Claude Code skills & agents that automate a full RTL/Verilog design flow - architecture, RTL coding, lint, simulation, self-checking testbenches, regression, debug, Vivado synthesis, and docs - driven from a one-page IP spec.

Claude Code skills + subagents for spec-driven analytical data products on AWS: Snowflake SQL transformations into Iceberg on Glue Catalog, liquibase-style DDL, config-driven ETL

Self-contained Claude skill to browse, scaffold & tfsec-scan SourceFuse ARC Terraform modules — no MCP server required

Turn the crank. The loop runs. A Claude Code skill for designing loops that prompt the agent instead of being the loop yourself.

A Claude Code skill: plan → codex verifies → build → codex verifies → fix. An independent verifier gates both the plan and the result.

Переиспользуемые скиллы для Claude Code. Первый — drawio: генерация редактируемых диаграмм draw.io / diagrams.net из текста.

Claude Code skill: design file-based agent loops (vision/specs/plan/prompts + bash engine) so you verify the output, not babysit every step. Single-machine, no special infra.

Provide Claude plugins with shared coding guidelines, review checks, and stack-specific skills to keep agents consistent across projects

One-command Claude Code power setup: orchestrator + PARA vault (Graphify-indexed markdown) + SDD agent factory + hooks + 8 plugins + 7 design skills

Codex / Claude Code skill for Douyin and Xiaohongshu content compliance review.

A local-first, MCP-native toolkit for building AI agents: engram (memory) + skillet (skills) + tracelet (tracing). One runnable end-to-end recipe.

Pixel-art virtual office that visualizes AI agent activity in real-time. Works with Claude Code, Gemini CLI, Codex, and any tool that can POST status.

Mission control for AI coding agents in tmux — a KDE Plasma 6 widget showing which Claude Code / aider / codex agent is working, waiting for input, or idle, with desktop notifications and one-key quick reply.

Claude Code × OpenAI Codex, as peers. A skill for cross-model collaboration (delegate / consult / roundtable) with a trace contract and eval suite.

⚔️ Prompt engineering, AI agents, workflows, automation, architecture reviews, and skill development for builders.

Detector-guided humanizer skill for Claude Code: lower the AI-detection score of English text while preserving every key term, number, citation, and the original logic. Keyless, agent-as-rewriter.