🤖 AI agents
AI coding agents, such as Claude Code, Codex or Cursor, can set up and configure Content Studio for you. Give them the documentation in a form they read well: an index and a full text of the docs, and an agent skill with the setup, the main APIs and the mistakes to avoid.
The documentation as text
The docs website publishes the documentation as plain Markdown, generated from the same pages you are reading:
/llms.txt: an index of every page with a short summary, in the llms.txt format./llms-full.txt: every page in one file.- Each page as Markdown: add
.mdto its address, such ashttps://djangocontentstudio.org/docs/edit-page.md.
Point your agent at https://djangocontentstudio.org/llms.txt, or paste a
page's Markdown address into the conversation.
The agent skill
The agent skill is a short guide for agents: how to install Content Studio, register models, lay out the edit page, pick widgets and formats, write conditional fields, and set up the dashboard, menus, extensions and analytics, with the mistakes the system checks report most. It links to the pages above for everything else.
The skill is published at
/skills/django-content-studio/SKILL.md.
For Claude Code, save it in your project, where Claude Code loads it when a
task involves Content Studio:
mkdir -p .claude/skills/django-content-studio
curl -o .claude/skills/django-content-studio/SKILL.md \
https://djangocontentstudio.org/skills/django-content-studio/SKILL.md
Other agents that support skills load it from their own skills folder. For
an agent without skills, add the file's address to your project's
AGENTS.md.
The skill and the text versions follow the current documentation. Download the skill again after you upgrade Content Studio.
Check the result
Agents make the same mistakes people do: a required field left off the edit page, a column that is not a field, two buttons with the same ID. Run the system checks after every change, and let the agent fix what they report:
python manage.py check