Materialize Agent Skills

Agent skills to help developers build with Materialize, a streaming database for real-time analytics. Agent skills are folders of instructions, scripts, and resources that AI agents like Claude Code, Cursor, GitHub Copilot, and others can discover and use to work more accurately and efficiently.

The skills in this repo follow the Agent Skills Open Standard.

Installation

npx skills add MaterializeInc/agent-skills

Once installed, skills activate automatically when your prompt matches their use case.

Available Skills

materialize-docs

Materialize documentation for SQL syntax, data ingestion, concepts, and best practices.

Use when:

• Writing or debugging Materialize SQL queries
• Setting up sources (Kafka, PostgreSQL, MySQL, webhooks, etc.)
• Creating materialized views, indexes, or sinks
• Configuring clusters or deployment settings
• Working with data ingestion or transformation patterns

Categories covered:

• SQL Commands (120+ command references)
• Core Concepts (clusters, sources, sinks, views, indexes)
• Data Ingestion (Kafka, PostgreSQL, MySQL, MongoDB, SQL Server, webhooks)
• Data Transformation (patterns, optimization, idiomatic SQL)
• Serving Results (sinks, BI tools, FDW)
• Integrations (CLI, client libraries, HTTP/WebSocket APIs)
• Security (RBAC, network policies, SSO)
• Self-Managed Deployments (AWS, Azure, GCP, Kubernetes)
• Management (monitoring, dbt, Terraform, disaster recovery)

Synced with the online Materialize documentation. Either one can be used to identify the Materialize version that introduced or updated a particular feature.

mcp-developer-analysis

Analyze a Materialize environment via the MCP Developer endpoint, and/or configure an MCP client (Claude Code, Cursor, VS Code, Zed, Continue, Windsurf, Claude Desktop) to connect to the materialize-developer server.

Use when:

• Checking environment health
• Investigating performance issues
• Troubleshooting stale materialized views
• Diagnosing memory pressure
• Auditing resource utilization
• Getting optimization recommendations
• Configuring an MCP client to connect to materialize-developer (Emulator, Cloud, or self-managed)
• Controlling which user or role the connection uses
• Switching between identities

Usage

Skills are automatically available once installed. The agent will use them when relevant tasks are detected.

Examples:

How do I create a Kafka source in Materialize?

Help me set up a materialized view that joins two sources

What's the syntax for CREATE SINK?

Skills can also be explicitly invoked in user prompts:

❯ /materialize-docs what sources are supported?

Compatibility

This skill follows the Agent Skills Open Standard and is compatible with 30+ AI agents, including:

• Claude Code
• Cursor
• GitHub Copilot
• Cline
• Windsurf

Skill Structure

Each skill follows the Agent Skills Open Standard:

skills/
  {skill-name}/
    SKILL.md              # Required: skill manifest with frontmatter
    README.md             # Optional: user-facing documentation
    references/           # Optional: curated reference material
    {topic}/              # Topic directories containing documentation
      index.md

Contributing

See CONTRIBUTING.md for details.

License

See LICENSE.

Changelog

• 2026-05-08: Add MCP client setup playbook to mcp-developer-analysis
• 2026-05-05: Rename mz-developer-analysis to mcp-developer-analysis
• 2026-04-28: Add mcp-developer-analysis skill
• 2026-02-09: Add materialize-docs skill