dial-your-context

Craft delta-based Instructions content for the Sanity Agent Context MCP.

4|1|Updated Feb 2, 2026
One-click install
npx skills add https://github.com/sanity-io/agent-context --skill dial-your-context
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: dial-your-context
Source: https://github.com/sanity-io/agent-context/tree/main/skills/dial-your-context
Command: npx skills add https://github.com/sanity-io/agent-context --skill dial-your-context

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Craft concise delta-based Instructions content for the Sanity Agent Context MCP, ensuring users update only what the schema cannot deduce and avoid duplicating existing guidance.

Core Features & Use Cases

  • Step-by-step workflow to help users craft precise, minimal Instructions that complement the generated schema.
  • Interactive guidance across seven steps: connect and set up, schema dialogue, define expected questions, verify findings, draft and verify instructions, deploy, and guardrail checks.
  • Outputs: a ready-to-deploy Instructions block and a GROQ filter for agent scope.

Quick Start

Initiate a session to craft delta-based Instructions tailored to your Sanity Agent Context MCP.

Frequently Asked Questions about dial-your-context

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I configure MCP instructions for Sanity when the agent misinterprets schema relationships?

Configuring MCP instructions for Sanity requires delta-based tuning to update only what the schema cannot deduce, enforcing constraints to correct agent misinterpretations of schema relationships.

What are MCP instructions and when do I need delta-based tuning for agent context?

MCP instructions are structured content guiding agent responses for Sanity data. Delta-based tuning for agent context is needed when the generated schema lacks implicit relationship rules or requires specific content filters.

How do I set up content filters and GROQ filters for Sanity agent scope?

Setting up content filters for Sanity agent scope involves crafting precise Instructions content and outputting a GROQ filter through a seven-step workflow including schema dialogue and guardrail checks.

Can I use this workflow to tune agent context without duplicating existing schema guidance?

Yes, you can tune agent context without duplicating existing schema guidance by crafting concise, delta-based Instructions that complement the generated schema and update only missing relationship information.

What's the best way to deploy a ready-to-deploy Instructions block for Sanity Agent Context MCP?

The best way to deploy an Instructions block for Sanity Agent Context MCP is following the interactive seven-step workflow: connect, schema dialogue, define questions, verify, draft, deploy, and execute guardrail checks.

Why does my Sanity agent return incorrect responses and how do guardrail checks help?

Sanity agents return incorrect responses when schema relationships are misinterpreted. Guardrail checks help by enforcing structured constraints during deployment to ensure safe, reliable instruction adherence.

Related Skills