---
title: "Personalizing Genie Code with instructions, skills, memory, and MCP"
description: "Genie Code adds personalization through persistent instructions, task-specific Agent Skills, shared repository conventions, and MCP servers. Instructions apply across every agent-mode session, while AGENTS.md and CLAUDE.md files can be discovered automatically for team conventions. Skills are markdown-based packages with names, descriptions, guidance, reusable code, or scripts, scoped to workflows and available personally or through Workspace/.assistant/skills/. MCP servers expose external tools, data, and workflows from managed services, Unity Catalog-connected servers, or Databricks Apps, with workspace admin controls and permission-based user selection. Managed OAuth flows for Google Drive, SharePoint, and GitHub are currently in beta, while the MCP marketplace and Genie Code settings support server discovery and setup."
---

# Personalizing Genie Code with instructions, skills, memory, and MCP

[Databricks](https://yomu.fyi/company/databricks) · Samantha Banchik, Gal Oshri, Romain Rigaux, Will Tipton, Chloe Chan · Jun 1, 2026

**Type:** Announcement

## Summary

Genie Code adds personalization through persistent instructions, task-specific Agent Skills, shared repository conventions, and MCP servers. Instructions apply across every agent-mode session, while AGENTS.md and CLAUDE.md files can be discovered automatically for team conventions. Skills are markdown-based packages with names, descriptions, guidance, reusable code, or scripts, scoped to workflows and available personally or through Workspace/.assistant/skills/. MCP servers expose external tools, data, and workflows from managed services, Unity Catalog-connected servers, or Databricks Apps, with workspace admin controls and permission-based user selection. Managed OAuth flows for Google Drive, SharePoint, and GitHub are currently in beta, while the MCP marketplace and Genie Code settings support server discovery and setup.

## Context

Teams need Genie Code to reflect coding standards, internal workflows, shared tools, and the context behind past decisions. Global instructions are useful for universal preferences, but task-specific context requires a more targeted mechanism, while external context may already reside in documentation systems, internal tools, and services.

## Approach / What changed

Use persistent custom instructions for universal preferences; AGENTS.md and CLAUDE.md for automatically discovered repository conventions; markdown-based personal or workspace Agent Skills for focused workflows; and approved MCP servers for governed access to external tools, data, and workflows. Workspace administrators control available servers and user permissions, with managed OAuth available in beta for selected services.

## Takeaways

- Agent Skills can include workflow guidance, reusable code, and executable scripts, with names and descriptions used to determine when a skill is relevant.
- Workspace skills stored in Workspace/.assistant/skills/ are available to everyone in the workspace and can standardize workflows such as incident response or data quality checks.
- MCP servers can be managed Databricks services, Unity Catalog-connected external servers, or custom servers hosted on Databricks Apps; users access approved servers they have permission to use.

**Tags:** [AI Agents](https://yomu.fyi/topic/ai-agents), [Databricks](https://yomu.fyi/topic/databricks), [Genie](https://yomu.fyi/topic/genie), [MCP](https://yomu.fyi/topic/mcp)

- Source: [Databricks](https://www.databricks.com/blog/personalizing-genie-code-instructions-skills-memory-and-mcp)
- Source URL: https://www.databricks.com/blog/personalizing-genie-code-instructions-skills-memory-and-mcp
- Ingested by Yomu: 2026-08-31T03:32:28.341Z

[Read original post](https://www.databricks.com/blog/personalizing-genie-code-instructions-skills-memory-and-mcp)