by Microck
An unappealing collection of Claude Skills and resources.
# Add to your Claude Code skills
git clone https://github.com/Microck/ordinary-claude-skillsGuides for using ai agents skills like ordinary-claude-skills.
Last scanned: 5/30/2026
{
"issues": [],
"status": "PASSED",
"scannedAt": "2026-05-30T15:31:10.591Z",
"npmAuditRan": true,
"pipAuditRan": true
}ordinary-claude-skills is an open-source ai agents skill for AI coding assistants such as Claude Code, Codex CLI, and ChatGPT, built by Microck. An unappealing collection of Claude Skills and resources. It has 257 GitHub stars.
Yes. ordinary-claude-skills passed SkillsLLM's automated security scan — a dependency vulnerability audit plus prompt-injection heuristics — with no high-severity issues. You can read the full report in the Security Report section on this page.
Clone the repository with "git clone https://github.com/Microck/ordinary-claude-skills" and add it to your Claude Code skills directory (see the Installation section above).
ordinary-claude-skills is primarily written in Python. It is open-source under Microck on GitHub, so you can review or fork the full source.
Yes. SkillsLLM lists many other AI Agents skills you can browse and compare side by side. Open the AI Agents category from the badge at the top of this page, or use the Related Skills and comparison links further down to weigh ordinary-claude-skills against similar tools.
No comments yet. Be the first to share your thoughts!
there are two ways to consume this library.
go to the static site. i have indexed everything with search and categories. it is much easier than digging through folders.
clone the repo to map these skills into your own mcp servers or agents.
clone the repo
git clone https://github.com/Microck/ordinary-claude-skills.git
cd ordinary-claude-skills
choose your weapon
custom skills, and upload the specific folder for the skill you want.verify
ask claude can you use the [skill name] skill now? if it says yes, you are gucci.
skills are basically fancy prompt packages and scripts that teach claude how to do specific things without you having to explain the context every single time. they load lazily (only when needed), which saves context window space and keeps claude from getting confused by instructions it doesn't need yet.
this repo aggregates hundreds of skills from anthropic, composiohq, k-dense-ai, and random internet geniuses.
i used to list all 600+ skills here, but it made the readme scroll for eternity.
view the full inventory on the documentation site →
categories include:
getting this to work depends on your environment. here is the recommended way to set things up if you are using mcp or a local client.
ordinary-claude-skills/
├── docs/ # the static website files
├── skills_all/ # everything
├── skills_categorized/ # everything in its right place
│ ├── backend/
│ │ └── api-design-principles/
│ └── web3-tools/
│ └── solidity-security/
└── README.md
if you are using a tool that requires a config file to point to skills, it usually looks something like this.
{
"mcpServers": {
"filesystem": {
"command": "npx",
"args": [
"-y",
"@modelcontextprotocol/server-filesystem",
"/path/to/ordinary-claude-skills/skills_all"
]
}
}
}
or map only the category or specific skill you need.
{
"mcpServers": {
"filesystem": {
"command": "npx",
"args": [
"-y",
"@modelcontextprotocol/server-filesystem",
"/path/to/ordinary-claude-skills/skills_categorized/[category]"
]
}
}
}
here is how you actually talk to claude once the skills are loaded.
load the debugging-strategies and frontend-design skills.
you:
i have a react component that is not rendering the list items correctly. please use the systematic debugging skill to analyze the code i paste next, and then use the frontend design skill to propose a fix.
claude:
acknowledged. i will apply the systematic debugging protocol. please paste the code.
load the competitive-ads-extractor skill.
you:
here is a url to a landing page. run the ads extractor and tell me what their primary value proposition is.
claude:
running extraction...
load the pdf skill.
you:
a client just sent me a scanned image of a spreadsheet pasted into a word doc and then exported as a pdf. i am losing my will to live. please use the pdf skill to extract the text so i don't walk the plank.
claude:
extracting text now. please drink some water while i handle this crime against data structures.
load the webapp-testing skill.
you:
i am about to push to prod on a friday afternoon. run the webapp testing skill on
localhost:3000and tell me if i am going to get fired.
claude:
starting playwright tests. i suggest you keep your resume updated just in case the login modal is broken again.
sometimes computers are hard.
claude refuses to use the skill: make sure you explicitly told claude the skill exists in the system prompt or that the file was successfully attached to the project context. usually it just doesn't know it's there.
"file too large" error:
some of these skills have massive dependency folders. ignore the node_modules inside skill folders. you only need the source scripts and the instructions.
skills contradicting each other:
don't load creative-writing and technical-documentation at the same time. claude will get confused about whether it should be shakespeare or a robot.
technically none for the repo itself, but individual skills have requirements.
python 3.x (for data analysis skills)node.js (for mcp builder and testing skills)playwright (if you want to do browser automation)i did not write most of these. i just collected them.
LICENSE file in each specific folder.credits go to anthropic, composiohq, k-dense-ai, and the other legends listed in the source tables. if you own one of these and want me to take it down, just open an issue and i will nuke it.