Traction scoreGitHub stars can be faked, so popularity alone can be misleading. Traction Score looks for broader signs of recent attention, adoption, and active maintenance.
This repository bundles 45 skills that teach an AI coding assistant how to build backend APIs in NestJS or Django, work with PostgreSQL and MongoDB, implement authentication, deploy to Cloudflare or GCP, and process multimedia and documents. It also includes systematic debugging, code review, chrome-devtools automation, and skill creation. It serves full-stack developers, AI engineers, and platform teams who need consistent domain knowledge across their AI-assisted workflow.
WHO IT'S FOR
UI/UX Designers
creating aesthetic interfaces with design principles
Full-Stack Developers
building production-ready backend and frontend systems
Compatible AgentsThe repository documents support for these agents. The skills may also work with other agents that can load SKILL.md files, but they may need some setup or small changes.
Compatibility not documented
Use this skill collection
Follow the documented setup, then try a first task.
Claude Code marketplace: ai-ml-tools
Installs: Installs the complete ai-ml-tools plugin bundle. The listed skillPaths are the positively supported mappings available in the marketplace manifest. · Claude Code
Before you start
Claude Code
For ai-multimodal or google-adk-python: Python and pip; configure Google AI Studio credentials or the documented Vertex AI project settings before using the relevant API.
Enter this slash command in a Claude Code session to add the ClaudeKit Skills marketplace.
/plugin marketplace add mrgoonie/claudekit-skills
Then enter this slash command in the Claude Code session to install the complete ai-ml-tools plugin bundle.
/plugin install ai-ml-tools@claudekit-skills
Before using ai-multimodal, install its Python SDK dependencies in the Python environment that will run the skill scripts.
pip install google-genai python-dotenv pillow
For Google AI Studio, obtain an API key from https://aistudio.google.com/apikey and set GEMINI_API_KEY in the process environment or one of the documented .env locations. For Vertex AI instead, set GEMINI_USE_VERTEX=true and VERTEX_PROJECT_ID to the project ID; VERTEX_LOCATION is optional.
Before using google-adk-python, install the documented stable Python package.
Installs: Installs the complete web-dev-tools plugin bundle. The listed skillPaths are the positively supported mappings available in the marketplace manifest. · Claude Code
Before you start
Claude Code
Enter this slash command in a Claude Code session to add the ClaudeKit Skills marketplace.
/plugin marketplace add mrgoonie/claudekit-skills
Then enter this slash command in the Claude Code session to install the complete web-dev-tools plugin bundle.
Installs: Installs the complete devops-tools plugin bundle. The listed skillPaths are the positively supported mappings available in the marketplace manifest. · Claude Code
Before you start
Claude Code
For Cloudflare workflows: Wrangler CLI and access to the relevant Cloudflare account; for Google Cloud workflows: gcloud CLI and project credentials.
For container workflows: Docker CLI and a running Docker daemon; for Kubernetes workflows: kubectl and access to the target cluster, with Helm when using chart workflows.
For database operations: the relevant MongoDB or PostgreSQL client and access credentials for the intended database.
For chrome-devtools scripts: Node.js, npm, Chrome and the documented platform libraries; ImageMagick is optional for screenshot compression.
Enter this slash command in a Claude Code session to add the ClaudeKit Skills marketplace.
/plugin marketplace add mrgoonie/claudekit-skills
Then enter this slash command in the Claude Code session to install the complete devops-tools plugin bundle.
/plugin install devops-tools@claudekit-skills
Before using chrome-devtools, locate its installed scripts directory. On Linux/WSL, run the supplied install-deps.sh there to install required Chrome libraries; macOS and Windows skip that system-library step. The source documents .claude/skills/chrome-devtools/scripts for the local collection layout.
In the installed chrome-devtools scripts directory, install its Node dependencies before running the scripts.
Installs: Clones the repository and copies its complete .claude contents into a chosen project's .claude directory. The skillPaths below are a bounded selection prioritized for the starter tasks, not the complete installed inventory. · Claude Code
Before you start
Claude Code
Git and a POSIX-compatible shell
An existing destination project .claude directory
For ai-multimodal or google-adk-python: Python and pip; configure Google AI Studio credentials or the documented Vertex AI project settings before using the relevant API.
For Cloudflare workflows: Wrangler CLI and access to the relevant Cloudflare account; for Google Cloud workflows: gcloud CLI and project credentials.
For container workflows: Docker CLI and a running Docker daemon; for Kubernetes workflows: kubectl and access to the target cluster, with Helm when using chart workflows.
For database operations: the relevant MongoDB or PostgreSQL client and access credentials for the intended database.
For chrome-devtools scripts: Node.js, npm, Chrome and the documented platform libraries; ImageMagick is optional for screenshot compression.
Platform: macOS/Linux or a compatible POSIX shell with git and cp
Run from: A directory in which the claudekit-skills repository may be cloned and from which the destination project path is reachable.
Run this terminal command from the chosen working directory to clone the repository.
Then run this terminal command, replacing the author-provided placeholder `/path/to/your/project` with the destination project's actual path. Keep the placeholder verbatim in the command until making that substitution.
Before using ai-multimodal, install its Python SDK dependencies in the Python environment that will run the skill scripts.
pip install google-genai python-dotenv pillow
For Google AI Studio, obtain an API key from https://aistudio.google.com/apikey and set GEMINI_API_KEY in the process environment or one of the documented .env locations. For Vertex AI instead, set GEMINI_USE_VERTEX=true and VERTEX_PROJECT_ID to the project ID; VERTEX_LOCATION is optional.
Before using google-adk-python, install the documented stable Python package.
pip install google-adk
Before using chrome-devtools, locate its installed scripts directory. On Linux/WSL, run the supplied install-deps.sh there to install required Chrome libraries; macOS and Windows skip that system-library step. The source documents .claude/skills/chrome-devtools/scripts for the local collection layout.
In the installed chrome-devtools scripts directory, install its Node dependencies before running the scripts.
Use the Inversion Exercise skill on this decision or problem: [describe the decision/problem]. Context and constraints: [provide the relevant context and non-negotiable constraints]. List the core assumptions, invert each one systematically, explore what each inversion would change, and identify which inversions are genuinely viable. Finish with a concise recommendation; do not implement any changes.
Use the Scale Game skill to review this architecture or approach: [describe the design]. Its expected workload is [volume, users, duration, speed, and failure-rate assumptions]. Pick the most relevant scale dimensions, test both 1000x smaller/faster/fewer and 1000x bigger/slower/more, and report what breaks, what survives, and the smallest validations we should run next. Do not modify or deploy anything.
Use the mermaidjs-v11 skill to create a Mermaid v11 flowchart for this process: [describe the steps, decisions, and participants]. Return one inline Mermaid Markdown code block with clear node labels and labeled decision branches. Do not render files or modify documentation yet.
