SanityAI4 MIN READ
12.11.2025
.md

The AI-Driven Journey: Crafting Content with Sanity MCP

Last updated: 17.03.2026

robot

Editor's Note

This entire blog post was created using AI through the Sanity MCP (Model Context Protocol) server. As you read about Sanity MCP, you're experiencing its capabilities in real-time. This article—originally written in December 2025 and updated in March 2026—demonstrates both the technology and the process that brought it to life, written by an AI assistant directly into the Sanity CMS through natural language commands.

The Meta-Story: Writing About AI With AI

Before we dive into Sanity MCP itself, let me share something remarkable: I, an AI assistant, am writing this blog post about AI-powered content management while using the very technology I'm describing. This is the power of Sanity MCP—a seamless integration between AI and content management that makes this meta-moment possible.

Here's what happened: A human asked me to create a blog post about Sanity MCP and explain that it was written by AI. I connected to the Sanity MCP server and wrote directly into the Sanity CMS through natural language. No manual copying and pasting. No Studio interface clicking. Just AI → Sanity → Your readers.

What is Sanity MCP?

Sanity MCP is a Model Context Protocol server that bridges AI assistants and your content management system. It went Generally Available in December 2025 and is now hosted remotely at mcp.sanity.io—no npm package to install, no local setup, just OAuth authentication and you're ready. It ships with 34 tools covering document operations, schema management, image generation, semantic search, and GROQ query execution.

Supported AI clients now include Claude Code, Cursor, VS Code with GitHub Copilot, Lovable, v0 (Vercel), Replit Agent, and OpenCode—making it the most widely adopted AI-to-CMS integration available.

The Technology Behind This Post

Step 1: Schema Deployment

Before AI can write to your Sanity CMS, the schema must be deployed:

1npx sanity@latest schema deploy

Step 2: MCP Configuration

As of Studio v4.21.0 (December 2025), the Sanity CLI auto-detects AI-powered editors and offers to configure MCP automatically when you run sanity init. For manual setup, create .vscode/mcp.json:

1{
2  "servers": {
3    "sanity-mcp": {
4      "url": "https://mcp.sanity.io",
5      "type": "http"
6    }
7  }
8}

Step 3: Natural Language Request

The developer simply asked: "Can you create a blog post about sanity mcp server and tell them how this blog post is written by AI". That's it. No forms, no Studio navigation—just a conversation.

Real-World Applications

Sanity Content Agent

Launched publicly in January 2026, the Sanity Content Agent is a schema-aware AI that operates at scale inside your Studio. It can audit thousands of documents, execute bulk edits across hundreds of documents simultaneously (staged for human review), and run pipeline transformations—turning a press release into a structured article complete with metadata and alt text. Early adopters include Morning Brew, CoinDesk, Braze, Complex, and Home Instead.

Agent Actions & Sanity Functions

The Agent Actions API lets developers trigger schema-aware LLM instructions from any environment—useful for automated translation pipelines, content enrichment, and cross-platform publishing without custom integrations. Sanity Functions provides a fully managed serverless runtime for these content lifecycle workflows, removing infrastructure overhead entirely.

Content at Scale

Marketing teams can ask AI to generate SEO-optimized landing pages, translate content to multiple languages while maintaining keyword structure, or identify content gaps by analysing trending topics—tasks that previously required weeks of manual effort.

The Human-AI Collaboration

Despite being written by AI, this post isn't replacing human content creators—it's augmenting them. The workflow is: Human Strategy → AI Execution via MCP → Human Review → AI Assistance for bulk edits → Human Approval and publish. The result is content that combines AI efficiency with human insight.

The Ethics and Transparency

Full disclosure matters. This post is proudly AI-generated. Just as photographers use cameras and writers use word processors, content creators can use AI. The tool doesn't diminish the craft—it enhances efficiency. Human review remains essential: AI draft → Human review → Published content. Transparency builds trust; hiding AI usage erodes it.

Getting Started With Sanity MCP

1# Install Sanity CLI and deploy schema
2npm install -g @sanity/cli
3npx sanity@latest schema deploy
4
5# Run sanity init — auto-detects your AI editor
6# and offers to configure MCP automatically

Once configured, try natural language commands like: "Show me my Sanity content model", "Query all blog posts from last month", or "Create a 1000-word technical post about X targeting developers with TypeScript examples". Specificity is key—vague prompts yield generic results.

The Future of Content Creation

MCP itself has grown into an industry standard: Anthropic donated the protocol to the Agentic AI Foundation (under the Linux Foundation) in December 2025, co-founded with Block and OpenAI. OpenAI adopted MCP in March 2025, and the March 2026 MCP Extensions spec lets developers layer capabilities on the stable core. Gartner predicts 40% of enterprise applications will feature task-specific AI agents by end of 2026, up from under 5% in 2025.

Content will be collaborative between humans and AI, speed will increase without sacrificing quality, and personalization will scale as AI generates audience-specific variations automatically.

Conclusion: A New Era of Content Management

Sanity MCP isn't about replacing writers. It's about empowering teams to create more, faster, and better—making content management conversational, intuitive, and intelligent. The future isn't AI versus humans. It's AI with humans, and it's already here.

References

Sanity MCP Server Documentation

Sanity Content Agent

Sanity Agent Actions

Model Context Protocol Specification 2025-11-25

Continue Reading
brewfather, mcp & ai
Home BrewingAI2 MIN

I Fed Brewfather's API to Claude and Built My Own MCP Server

Brewfather is a comprehensive app for homebrewers that manages recipe design, batch tracking, fermentation logs, water chemistry, and ingredient inventory, making it a central hub for homebrew recipes. An MCP (Model Context Protocol) server acts like a plugin that exposes local tools and data to an AI such as Claude, letting it call a locally run server without any cloud intermediary or custom UI. Using Brewfather’s REST API, the author fed the documentation to Claude and quickly generated a local MCP server that connects Brewfather to Claude Code. This setup enables three key capabilities: browsing and inspecting real Brewfather recipes, checking the brew schedule and batch stages, and verifying whether the ingredient inventory is sufficient for an upcoming brew day. The main benefit is turning brewing data into a conversational interface, so users can simply ask questions like whether they are ready to brew on a given day. Future extensions could include logging fermentation readings, auto-creating batches from recipes, and sending notifications as batches progress through stages.

Read post
Ai genreated image wat workflow
SanityAI3 MIN

Using the WAT Framework: Writing Sanity MCP Workflows That Make Claude Consistent and Reliable

The text explains why open-ended AI instructions like “write a blog post about TypeScript” lead to inconsistent results. Because models are probabilistic, they vary structure, miss fields, and overlook edge cases, which is problematic for repetitive, structured tasks such as publishing to a CMS. To solve this, it introduces the WAT framework: Workflows, Agents, Tools. Workflows are plain-language markdown SOPs that encode domain knowledge and specify steps. Agents (Claude) handle reasoning and decisions. Tools are deterministic scripts or APIs, like the Sanity MCP, that execute actions. This separation narrows the decision space and keeps behavior consistent across sessions. A concrete example is the draft_blog_post workflow, which fetches authors and categories from Sanity, requires outline approval, and strictly defines document shape and constraints, including a 5000-byte body limit. Workflows evolve through a self-improvement loop: each failure adds new rules and edge cases. To get started, you document repeatable tasks, inputs, tools, steps, and edge cases, store them in .claude/wat/workflows/, and reuse them for faster, cheaper, and more reliable AI-assisted work.

Read post