Prismic

Your Prismic content library becomes your AI agent's brain

Customers ask about your products, features, or policies. Your AI agent searches your Prismic CMS in real time, retrieves the exact content slice, and delivers accurate answers grounded in your published content. No stale knowledge base. Always up to date.

Chosen by 800+ global brands across industries

CMS-powered answers on demand

Your AI agent taps directly into Prismic's content API, fetching documents, searching by keyword, and filtering by type or tag to answer every visitor question with authoritative content.

Prismic

Use Cases

Content-driven conversations

See how businesses use their Prismic CMS as the knowledge backbone for AI-powered customer conversations that always reflect the latest published content.

Answering Product Questions from Your CMS

A prospect asks 'What features does your enterprise plan include?' Your AI Agent searches your Prismic repository for documents tagged 'pricing' and type 'feature_page', extracts the relevant content slices, and responds with an accurate, up-to-date feature breakdown. The visitor gets a detailed answer without waiting for sales. Your content team controls the source of truth.

Self-Service Knowledge Base Without a Separate Tool

A customer messages 'How do I set up SSO?' Instead of redirecting them to a help center, your AI Agent performs a full-text search in Prismic for 'SSO setup', finds the matching documentation page, and walks the customer through the steps conversationally. Your existing CMS doubles as a dynamic knowledge base, no separate system needed.

Multilingual Visitors Served in Their Language

A French-speaking visitor asks a question on your site. Your AI Agent detects the language, queries Prismic with the 'fr-fr' locale parameter, and retrieves the localized version of your content. The visitor receives answers in their native language, pulled directly from your translated Prismic documents. Global support from day one.

Try
Prismic

Prismic

FAQs

Frequently Asked Questions

How does the AI agent search my Prismic content in real time?

The agent uses Prismic's Content API full-text search predicate to query your entire repository by keyword. When a visitor asks a question, the agent constructs a search query, retrieves matching documents with their content fields, and synthesizes a conversational answer. All queries run against your live published content using the master ref.

Can the agent serve content from different Prismic content types like blog posts, FAQ pages, or landing pages?

Yes. The agent uses the 'Get Documents by Type' endpoint to filter content by any custom type defined in your Prismic repository. Whether it is a blog_post, faq_item, or product_page, the agent retrieves only the relevant document type for each question.

What Prismic credentials does Tars need to connect?

Tars requires your Prismic repository name (subdomain) and a permanent access token. You generate the token in your Prismic repository's API settings under the 'Permanent access tokens' section. This grants read-only API access. Tars never accesses your Prismic writing API or admin settings.

Does the agent handle multilingual Prismic content?

Yes. The agent passes a language code parameter (like 'en-us' or 'fr-fr') when querying Prismic. If your repository has localized document variants, the agent retrieves content in the visitor's detected language. All of Prismic's supported locales work with this integration.

Does Tars store copies of my Prismic documents?

No. Tars queries your Prismic repository in real time during each conversation. Document content is fetched live and used only to generate the response. There is no separate copy or cache of your CMS data maintained on Tars servers.

How is this different from using Prismic's built-in search on my website?

Prismic's built-in search returns a list of document links. Tars AI Agent understands the visitor's intent, searches your repository contextually, and provides a conversational answer that synthesizes information from multiple documents. It also works on WhatsApp, SMS, and other channels beyond your website.

What happens when I update content in Prismic after the agent is live?

The agent always queries against your current master ref, which points to the latest published version of your content. When you publish changes in Prismic, the agent immediately starts returning the updated content. There is no delay or retraining required.

Can the agent retrieve content from Prismic integration fields and linked documents?

Yes. The agent supports Prismic's fetchLinks parameter and integration fields ref, allowing it to pull content from linked documents and external data sources connected to your Prismic repository. Related author bios, category metadata, and cross-referenced pages are all accessible.

How to add Tools to your AI Agent

Supercharge your AI Agent with Tool Integrations

Don't limit your AI Agent to basic conversations. Watch how to configure and add powerful tools making your agent smarter and more functional.

Privacy & Security

We’ll never let you lose sleep over privacy and security concerns

At Tars, we take privacy and security very seriously. We are compliant with GDPR, ISO, SOC 2, and HIPAA.

GDPR
ISO
SOC 2
HIPAA

Still scrolling? We both know you're interested.

Let's chat about AI Agents the old-fashioned way. Get a demo tailored to your requirements.

Schedule a Demo