From IonBin New tools and workflows every week. See what’s new
Package your best prompts as reusable agent skills
SOPs & process documentation · Dev, web & IT · checked 2026-10-17
Agent Skills are folders with instructions and resources that Claude and Codex load when relevant. Turn a process you repeat into a skill, test it, and share it with your team.
Independent review. IonBin has no partnership with any tool on this page and was not paid for this page. Our evaluation is our own, based on public documentation and our research. Product names are trademarks of their owners and are used for identification only. How disclosure works.
Goal
Stop re-pasting long instructions; let agents load the right playbook automatically.
Connects through: MCP
Steps
#
Who
What happens
How
Status
1
You
Pick a repeatable process and write the steps, examples and any scripts into a SKILL.md folder
Review skill outputs and update the skill, not just the output
Human step
Confirmed
Confirmed means the vendor documents it (linked). It doesn't mean IonBin has tested it. AI-platform and human steps need no source.
Human approval
You review outputs and own the skill contents.
Watch out for
Keep the human approval step: You review outputs and own the skill contents.
Starter prompt
Goal: Stop re-pasting long instructions; let agents load the right playbook automatically. Tools involved: Claude, Codex, GitHub. Your part: loads the skill when relevant, from ~/.claude/skills or a project’s .claude/skills in claude code; loads skills in the same skill.md format from .agents/skills in the repository; versions the skills library; changes go through pull requests. Work step by step, show me each draft or proposed change, and wait for my approval before anything is sent, published, signed, filed or changed in a live system.
Assembled from this page’s goal and steps. Paste it into your AI platform and adjust it to your situation.
Every MCP and API step is documented for this platform.
Perplexity
To verify
GitHub: pairing not named in vendor docs
Computed from the MCP and API steps only: "Confirmed" needs the tool's docs to name the platform and the platform to support the connection. Browser, upload and human steps work from any platform.
What we checked (2026-10-17)
The format. A skill is a folder with a SKILL.md file that has at least a name and a description, plus optional scripts and reference files. Agent Skills is published as an open format at agentskills.io.
Claude. Custom skills work in Claude Code (~/.claude/skills/ for personal skills, .claude/skills/ for a project), in claude.ai settings and through the Claude API. Anthropic says custom skills don’t sync across these surfaces, so upload them to each one you use. It also says to use skills only from sources you trust, because a skill can contain code.
Codex. Codex reads skills from .agents/skills folders, from your working directory up to the repository root, as well as user and admin locations. It loads a skill’s full instructions only when it picks the skill. You can also call one directly with $ or /skills.
Portability. Both use SKILL.md, but the folder locations differ. Keep one source folder and copy or link it into each location.
Set it up
Choose a task you’ve explained more than three times.
Write the steps, one good example and the common mistakes.
Test it on a real task in each agent.
Store the library in a GitHub repository, so changes get reviewed.
