Prismic
Prismic is a headless CMS that enables developers and content teams to manage and deliver content seamlessly across various platforms.
Verdict
Common use cases
- Audit published blog posts by date
- Draft new content from existing page structure
- Search documentation for specific terms
- Generate content inventory reports
- Pull integration field data for analytics
Integration
- Vendor
- Prismic
- Category
- other
- Auth
- API_KEY
- Tools
- 13
- Composio slug
prismic
Tools
- Content API: Get Documents with Integration Fields
Tool to fetch prismic documents including integration fields data. use when you need both core content and integration fields in one query. call after obtaining a valid ref token.
- Get Ordered Documents
Tool to fetch prismic documents ordered by a specified field. use after obtaining a valid api ref when you need sorted document results, e.g., orderings='[my.article.date desc]'.
- Get Prismic Document by ID
Tool to retrieve a specific prismic document by its id. use after obtaining a content ref.
- Get Prismic Documents by Tag
Tool to fetch prismic documents by tag. use when retrieving all documents associated with a specific tag.
- Get Prismic Documents by Type
Tool to fetch prismic documents by type. use when you need to retrieve documents with a specific custom type after obtaining the repository's master ref.
- Get Prismic Documents with Fulltext Search
Tool to fetch prismic documents using a full-text search predicate. use after obtaining the repository's master ref when you want to search for specific terms within documents.
- Get Prismic Documents with Predicates
Tool to fetch prismic documents using specified predicates. use when filtering documents based on multiple conditions after obtaining the repository's master ref.
- Get Prismic Repository Info
Tool to retrieve prismic repository metadata. use after authentication to load refs, types, languages, and other repository settings.
- Get Prismic Repository Refs
Tool to retrieve a list of repository refs (master and releases). use when you need the current repository references before executing any content queries.
- Get Prismic Tags
Tool to retrieve all tags from prismic repository. use when you need the full list of tags for filtering or categorization.
- Prismic: Get Custom Types
Tool to retrieve all custom types and slices from the prismic repository. use when you need to list custom types definitions after authentication.
- Prismic: Get Documents by Custom Field
Tool to fetch prismic documents by custom field predicate. use after obtaining a repository ref when filtering content by field constraints.
- Prismic: Query Documents
Tool to query prismic documents using predicates and pagination. use when you need to fetch multiple documents from a repository after obtaining a ref.
Setup
Setup guide
- 11. Open your Switchy workspace and navigate to Settings > Integrations > Add MCP. 2. Select Prismic from the list and click Connect. 3. In a new tab, log into your Prismic dashboard and go to Settings > API & Security. 4. Copy your repository's Permanent Access Token (or generate a new one if none exists). 5. Paste the token into Switchy's API Key field and click Authorize. 6. Switchy will fetch your repository metadata to confirm the connection works. 7. Open any Space and type '@Prismic get repository info' to test — you should see your repo's master ref, custom types, and available languages. 8. To query documents, first run '@Prismic get repository info' to grab the current ref, then use that ref in follow-up queries like '@Prismic get documents by type' or '@Prismic search documents for [term]'.
What teammates see: by default, memories from Prismic are scoped to the Space (PROJECT visibility) - you can mark any memory PRIVATE or share it ORG-wide.
Works well with
Top models
Compatibility data appears once enough Spaces have used this MCP together with a given model.
How Switchy teams use it
Starter prompts
List Recent Blog Posts
@Prismic get the repository ref, then fetch the last 10 documents of type 'blog_post' ordered by publication date descending.Open in a Space →
Search Help Docs
@Prismic get the repository ref, then search all documents for 'password reset' and return the top 5 matches with their titles and IDs.Open in a Space →
Audit Tagged Content
@Prismic get the repository ref, then fetch all documents tagged 'product-launch-2024' and list their titles and types.Open in a Space →
Pull Page by ID
@Prismic get the repository ref, then retrieve the document with ID 'ZXy12AbCdEfG' and show me its title, type, and all field data.Open in a Space →
Generate Content Inventory
@Prismic get the repository ref, then fetch all documents of type 'landing_page' and return their titles, last publication dates, and URLs.Open in a Space →
Example outputs
Illustrative - representative of the model's voice and quality, not literal recordings.
@prismic search all blog posts published in the last 30 days and list them by date, newest first
Found 12 blog posts published in the last 30 days. Here they are ordered by publication date: 1. "Migrating to Headless CMS" (Dec 15, 2024) 2. "API-First Content Strategy" (Dec 10, 2024) 3. "Building with Slices" (Dec 3, 2024) [...9 more entries] Each post includes title, publication date, author, and excerpt. This example illustrates how the MCP combines date filtering with ordering predicates to return sorted results from your Prismic repository.
This example uses Get Ordered Documents with date-range predicates. Requires a valid API key with read access to your repository. The MCP fetches the master ref automatically, then applies filters—no manual ref management needed. Results reflect your repository's actual content structure and custom type definitions.
@prismic find the document with ID 'ZXk9aBAAACEAw2Tn' and show me its integration fields data
Retrieved document "Product Launch Announcement" (type: article, ID: ZXk9aBAAACEAw2Tn). Core content: Title, body slices, featured image, publication date. Integration fields detected: - Shopify product IDs: ["gid://shopify/Product/123", "gid://shopify/Product/456"] - Analytics tags: ["launch", "q4-2024", "featured"] This example shows how the MCP surfaces both standard content fields and integration field data in a single query, useful when your Prismic documents reference external systems.
This example uses Get Documents with Integration Fields to retrieve linked data from external platforms. Integration fields must be configured in your Prismic repository settings first. The MCP returns raw integration field values—you'll need separate API calls to the linked systems (Shopify, etc.) to resolve full details.
@prismic search for 'accessibility' across all document types and summarize the main themes in our content
Full-text search for "accessibility" returned 8 documents across 3 types: - 4 blog posts discussing WCAG compliance, screen reader testing, and keyboard navigation - 3 product pages highlighting accessible design features - 1 landing page for an accessibility audit service Common themes: Most content emphasizes WCAG 2.1 AA standards, with recurring mentions of color contrast, semantic HTML, and ARIA labels. The blog posts skew technical (developer audience), while product pages focus on end-user benefits. This illustrates how the MCP's full-text search combines with AI synthesis to surface content patterns.
This example pairs Get Prismic Documents with Fulltext Search with AI summarization. The search scans all text fields in your repository—results depend on your content volume and how you've structured custom types. No write access needed, but large repositories may hit rate limits on complex queries.
Use-case deep-dives
When Prismic MCP beats manual CMS checks for content teams
A 3-person marketing team publishes blog posts and landing pages through Prismic twice a week. The MCP shines here because the content lead can ask Switchy to pull all draft articles tagged 'Q1-launch', check publication dates, and flag missing SEO fields—without opening the Prismic dashboard. The Get Documents by Tag and Fulltext Search tools handle 90% of pre-publish audits. The trade-off: if your team publishes more than 20 pieces daily, the 13-tool scope starts to feel narrow (no bulk-update or workflow-state tools). For small teams shipping predictable content calendars, this MCP turns Switchy into a lightweight editorial assistant that lives where your team already works.
Prismic MCP for support teams with structured help content
A 5-person support team maintains 200+ help articles in Prismic, organized by product area and customer tier. The MCP works when agents need fast answers mid-ticket: ask Switchy to search help docs for 'API rate limits', get the current article text, and paste it into the reply. The Fulltext Search and Get by Type tools make this instant. The boundary: Prismic's content API returns full document JSON, so if your articles embed complex slices or media, the MCP output gets verbose. This setup wins when your knowledge base is text-heavy and agents need retrieval speed over layout preview. For teams fielding 50+ tickets daily with Prismic as the single source of truth, the MCP cuts lookup time in half.
When Prismic MCP speeds up multi-client content reporting
A 6-person agency manages Prismic repositories for 12 clients, each with 50-500 pages. Monthly audits check for outdated content, broken integration fields, and missing translations. The MCP handles this: connect Switchy to a client's repo, pull all documents ordered by last-publish date, filter by language, and export a summary. The Get Ordered Documents and Integration Fields tools make cross-repo audits scriptable without custom code. The catch: API key auth means each client needs a separate Switchy connection, so if you're juggling 20+ clients, key rotation becomes overhead. For agencies with stable client rosters and predictable audit cadences, this MCP turns a 2-hour manual task into a 10-minute Switchy query.
Frequently asked
What does the Prismic MCP do in Switchy?
It connects Switchy to your Prismic repository so AI can read and search your content. The MCP fetches documents by type, tag, or ID, runs full-text searches, and pulls integration field data. Use it when you want AI to answer questions about your published content or draft new pages that match your existing structure.
Do I need admin access to connect Prismic?
You need a Prismic API key with read access to the repository. Permanent API keys work best because they don't expire. Repository tokens are scoped to a single repo, so if you manage multiple Prismic projects you'll need separate connections in Switchy. Admin rights aren't required—just the ability to generate or retrieve the key.
Can the Prismic MCP create or edit documents?
No. The MCP is read-only. It fetches published and draft content but can't write back to Prismic. If AI drafts a new page, you'll copy the output into Prismic manually. This keeps your content workflow safe—AI suggests, humans approve and publish.
Why use this instead of Prismic's REST API directly?
The MCP handles ref tokens, predicate syntax, and pagination automatically. AI can query your content in plain English without writing API calls. If you already have custom scripts that work, keep them. Use the MCP when you want conversational access to your repository without maintaining code.
Who on the team should connect Prismic to Switchy?
Whoever manages your Prismic repository and can generate an API key. Once connected, everyone in your Switchy workspace can ask AI to search or summarise your content. The connection doesn't count against Prismic's plan limits—it's just another API client reading your published data.