Skip to main content

AI agents (MCP clients)

This page covers external AI agents — Claude, ChatGPT, Cursor, and custom clients connected via MCP.

For in-platform specialist agents (enrichment, SEO, taxonomy, approvals in the Agent Control Center), see Agentic PIM and Specialist agents.

MCP agents vs specialist agents

Specialist agents (ACC)MCP agents (this page)
WhereAgent Control Center in PIMExternal AI clients
RunsScheduled, event-triggered, missionsUser-driven tool calls in chat/IDE
WritesApproval queue → apply on approveNamed MCP tool calls, auditable via API
QualityInsights + Catalog QualitySame catalog; use tools to query and act

An MCP agent is an AI client connected to CataZenta via MCP with a goal, tools, and guardrails — not a single chat message.

What agents can run

Agent missionTool chain (examples)
Catalog auditorfind_products_needing_enrichmentanalyze_product_completeness → create tasks
Publish botlist_connectionsquick_publishget_export_job_items
Content writerget_productget_product_attributesset_product_attribute
Taxonomy botlist_categoriesupdate_product assignments
Quality analystAgent insights and channel readiness before publish (see Catalog quality)

Design principles

  1. Structured data first — agents read families and attributes, not PDFs.
  2. Explicit actions — every write is a named tool call, auditable via API.
  3. Human gates — agent proposals use the approval queue; use PIM workflows for legal/brand sign-off before active + publish.
  4. Idempotent jobs — syndication via export jobs, not ad-hoc HTTP hacks.

Get started

  1. Agentic PIM overview — in-platform specialist agents
  2. MCP quickstart
  3. Tool reference
  4. Agent workflow patterns

Enterprise teams often pair agents with service accountsrecipe.