Kontent.ai

@Kontent.ai list all content types in our production environment so I can see what structured content models we have available

Retrieved 8 content types from your Kontent.ai environment: Article (fields: title, body, author, publish_date), Product Page (fields: name, description, price, images), Landing Page (fields: hero_title, hero_image, sections), Author Bio (fields: name, photo, bio_text), FAQ Item (fields: question, answer, category), Case Study (fields: client_name, challenge, solution, results), Newsletter (fields: subject, preview_text, content_blocks), and Press Release (fields: headline, date, body, contacts). Each type defines the structure and field constraints for that content category.

This example shows how the MCP surfaces your headless CMS schema. Useful for understanding what content structures exist before creating or querying items. Requires read access to the Management API; the list can be long in mature projects, so consider filtering by specific type names if you know what you're looking for.

@Kontent.ai get the Spanish language variant of our 'Q2 Product Launch' article so I can review the translated version

Retrieved Spanish (es-ES) variant of content item 'Q2 Product Launch': Title is 'Lanzamiento de Producto Q2', body contains 847 words of translated marketing copy, publish date set to 2024-06-15, and workflow step shows 'Ready for Review'. The variant includes 3 linked assets (product screenshots localized for Spanish market) and references 2 related product pages also available in Spanish.

Demonstrates language variant retrieval, a core feature for multilingual content workflows. You must know both the item identifier and the language codename. If the variant doesn't exist for that language, the MCP returns an error—check available languages first with the List Languages tool to avoid failed requests.

@Kontent.ai fetch all published articles from the last 30 days and summarize the main topics we've been covering

Found 12 published articles in the past 30 days. Dominant themes: 5 articles on AI integration features (focusing on workflow automation and content generation), 4 on customer success stories (enterprise clients in finance and healthcare sectors), 2 on product updates (new API endpoints and webhook improvements), and 1 on company culture (remote work practices). Average article length is 1,200 words. All articles include author attribution and are tagged for SEO with 3-5 keywords each.

This synthesis example combines the List Content Items tool with AI reasoning to extract patterns. The MCP provides raw content data; the AI interprets themes and trends. Pagination may be needed for high-volume environments—responses include continuation tokens if more items exist beyond the initial batch. Delivery API rate limits apply.

When Kontent.ai wins for multi-language product releases

A 6-person product marketing team ships feature announcements in 8 languages every quarter. They need to verify translations are live before sending emails to regional lists. The Kontent.ai MCP is the right call here because it surfaces language variants directly in Switchy—no context-switching to the CMS dashboard. One person drafts the standup prompt that checks all 8 variants for a given content item, confirms publication status, and flags missing translations. The team runs it every launch morning. This works cleanly when your content model is stable and item IDs are predictable. If you're constantly renaming items or your taxonomy shifts weekly, the identifier lookups get brittle. For teams shipping localized content on a schedule, this MCP turns a 20-minute manual check into a 30-second Switchy run.

Using Kontent.ai to inventory published content types

A 3-person legal ops team at a fintech needs to audit all customer-facing content for regulatory disclosures before a quarterly filing. They use the List Content Types and List Content Items tools to pull a full inventory of published articles, FAQs, and landing pages from Kontent.ai. The MCP is a good fit because it exposes the delivery API in Switchy, so they can filter by content type and check metadata without exporting CSVs. One person writes the prompt once, the team re-runs it each quarter. The limitation: if your content model has 50+ types or you need deep taxonomy traversal, the pagination gets slow and the prompt gets messy. For compliance checks on a stable content model with under 20 types, this MCP saves the export-and-spreadsheet dance.

When Kontent.ai helps customer support find KB articles

A 4-person support team fields tickets about product features and needs to link customers to the right help center article. They use the List Content Items tool to search Kontent.ai's knowledge base by keyword and surface the correct article URL in Switchy. This works when your KB lives in Kontent.ai and your team already uses Switchy for ticket triage. The MCP is borderline if your search volume is high—Kontent.ai's delivery API isn't optimized for fuzzy search, so you'll get better results with a dedicated search MCP if you're doing 100+ lookups a day. For small teams doing occasional KB lookups during ticket review, the Kontent.ai MCP keeps the workflow in one place without adding another search tool to the stack.