MCP & AI Queue
12 articles
- What is the Elovate MCP server?What the Elovate MCP server is, what an AI assistant can and cannot do with it, and the rules that keep you in control.
- What the Elovate MCP can and can't doA complete overview of what an AI assistant can do through the Elovate MCP, what always needs a person in Elovate, and what is never possible.
- Connecting Claude or another AI assistant to ElovateHow to get access to the Elovate MCP server and connect it to Claude or another MCP-compatible assistant.
- What can I ask? Example prompts for the Elovate MCPPractical example prompts for analysing your catalogue, proposing changes, building workflows and reviewing AI output through MCP.
- Working with the AI queue: approve, decline and revertHow AI proposals land in the AI queue, what each status means, and how approving, declining and reverting work.
- Proposals vs workflows: which one to useWhen to let the AI propose a direct change in the AI queue, and when to set up an AI workflow instead.
- Roles and permissions with MCPWhat your Elovate user rights mean when you work through MCP, and why reading is open but deciding is not.
- Analysing your catalogue with MCPUse the Elovate MCP to measure content quality, find gaps and decide where AI enrichment will pay off most.
- Setting up and testing a workflow through MCPHow the assistant builds an AI workflow, tests its cards and tunes the prompt, and what to know about chains and moderation.
- AI generation with Elovate via MCP: what's the difference?How AI generation through MCP and the AI queue differs from exporting, prompting and re-importing your product data.
- Elovate MCP vs building your own AI toolWhat Elovate already handles underneath the MCP, and what tends to go wrong with a self-built or vibe-coded AI integration.
- Troubleshooting the Elovate MCPCommon reasons why a proposal is refused, a workflow step doesn't run or a tool is missing, and how to fix them.