- DRAFT
- PUBLISHED
- METADATA
- REVIEW
Short answer
How can I automate Plone publishing with AI?
You can use text commands to trigger workflow transitions, create new content, or update metadata fields. Your AI handles the specific Plone actions while you manage the editorial decisions. This setup gives you direct control over the publication pipeline.
Content destinations
Where your content lands.
The AI moves data between these specific states based on your instructions.
DRAFT
New content created
The AI uses create_content to spin up new entries. You get a fresh document ready for review.
PUBLISHED
Live on site
Your agent triggers transitions to move items from draft to published. The content goes live immediately.
METADATA
Updated info
The AI modifies existing fields via update_content. This keeps your tags and descriptions current.
REVIEW
Human oversight
The AI flags items that need your eyes. You decide when a document is actually ready for the next stage.
The workflow
What your AI does when content arrives.
The AI acts as a bridge between your chat interface and the Plone backend.
Fetch existing data
Your agent looks up the current state and metadata of a specific document to understand what needs changing.
get_contentget_workflowModify document details
The AI updates specific fields like titles, descriptions, or custom metadata based on your text input.
update_contentAdvance the lifecycle
The AI moves a document through its defined workflow, such as moving a post from 'Draft' to 'Review'.
transition_workflowManage the library
The AI can generate entirely new entries or remove old ones from the system.
create_contentdelete_content
Try it
Copy these to start.
Use these prompts to begin interacting with your Plone instance.
Starting points
Replace the bracketed text with your actual Plone document IDs or specific field names.
Plone Connector
You're all set. Choose your MCP client and follow the setup instructions.
https://edge.vinkius.com/vk_preview_g6iXcoVFJw211BEVIrauCIn8dk9jo3oKnobRjnbL/mcpClaude Desktop
Follow the steps below to connect in seconds.
- 1In Claude Desktop, open Settings → Connectors.
- 2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
- 3Click Add and start a new chat — Plone capabilities are ready to use.
{
"mcpServers": {
"plone-alternative-mcp": {
"url": "https://edge.vinkius.com/vk_preview_g6iXcoVFJw211BEVIrauCIn8dk9jo3oKnobRjnbL/mcp"
}
}
}Create a new news article in Plone titled 'Annual Report' with the description '2023 summary'.
Move the document with ID 123 from 'Draft' to 'Published'.
Update the metadata for the document 'Project Alpha' to set the category to 'Internal'.
Check the current workflow state for the document titled 'Summer Campaign'.
Claude
ChatGPT
Cursor
VS Code
Windsurf
Claude Code
JetBrains
Cline
Start here
Connect Plone once, then ask.
Link your Plone instance to Vinkius in one step. Your credentials stay encrypted, and you can immediately start giving commands to your AI client.
Connect Plone to your AIFAQ
How the Plone Connector behaves
- 01
Can the AI write the actual article text for me?
No. The AI uses the tools to create the content object and update its fields, but you must provide the actual text you want to publish.
- 02
Will the AI delete content without asking?
The AI only deletes content if you explicitly command it to do so using a delete command.
- 03
Can the AI change the site's theme or layout?
No. The tools are limited to content creation, metadata updates, and workflow transitions.
- 04
How does the AI know which workflow stages exist?
The AI uses the get_workflow tool to see the available transitions for a specific piece of content.
- 05
Can the AI bypass the human review stage?
The AI can trigger a transition to 'Published' if your Plone workflow allows it, but it cannot change the rules of the workflow itself.
More questions about Plone? See everything the Plone Connector can do
Connect Plone to Claude, Cursor, ChatGPT & more
