Run your SOP inside Claude

Last updated: October 3, 2026

A written SOP sitting in a doc is a reference. The same SOP handed to Claude is something that can do the work with you. The difference is not the writing, it is whether the model has the process in front of it instead of guessing at how your business does things.

What can Claude read?

Claude supports Agent Skills: a folder with a SKILL.md file inside it, holding a name, a description and your instructions in plain Markdown. Once a skill is available, Claude pulls it in when the work matches the description, so you are not pasting the process into the chat every time. Skills work across the Claude apps, Claude Code and the Agent SDK.

That is the format Edessa exports to, which means a workflow you recorded is already in the shape Claude wants.

The details move, so check Anthropic’s skills docs rather than a blog post. The install steps were checked against the vendor docs on October 3, 2026.

Where does the file go?

Export the workflow as a Skill. You get a zip holding one folder named after the workflow, with SKILL.md and any attached files inside it.
On claude.ai, make sure code execution is on in Settings, then open Customize, then Skills, click +, choose Create skill, then Upload a skill, and pick the zip. In Claude Code, unzip it into your skills folder: unzip -o ~/Downloads/<skill>.zip -d ~/.claude/skills/
Ask for the task by name. On claude.ai it gets used automatically when a chat matches the workflow. If Claude does not reach for it, the description is too vague, not the process.

Try Edessa free

Three asks worth trying

Once the process is loaded, the useful prompts are not “summarise this.” They are the ones that use the fact that Claude can now see how your business actually works.

"Run this process for the client I just added, and stop before anything gets sent." Asking it to stop gets the work done and keeps the last call yours.
"Where in this process am I doing something a tool could do?" Claude can see the whole sequence at once, which you cannot while you are in it. Expect two or three obvious answers and one you had not thought about.
"A new hire is doing this tomorrow. What will they get wrong?" This finds the gaps a written SOP hides: the step that assumes context nobody wrote down.

How do you keep it current?

This is where documentation goes stale. The process changes, the file does not, and six months later the skill is confidently wrong. Keeping it right by hand means editing a Markdown file every time a step changes.

The alternative is re-recording. Do the task once the new way, replace the file with the new export, and the thing your AI tool reads matches what you actually do. That takes about as long as the task does.

The recording method is here, including the manual version if you would rather not use a tool for it.

Who wrote this. Thaddeus Demeke, Founder of Elevaite and the founder of Edessa. Five years at Genentech, most recently as a Senior Data Scientist, then building AI systems inside small businesses across agencies, home services, real estate and ecommerce. I run my own processes this way.

See how the AI SOP generator works

Run your SOP inside Claude

What can Claude read?

Where does the file go?

Three asks worth trying

How do you keep it current?

Related reading