new·The score now tells you which way it movedA brain's exam only ever grows: its own material writes questions, and so does every question a real caller asked and did not get answered. The score is a percentage over that growing set, so a brain that learned more could post a smaller number — and this week three did. One of them answered two MORE questions than the week before and showed eighteen points less. Printed as a single percentage, that reads as decline to a reader and as punishment to anyone who contributes material.all news →
mozg.beta
Sign in

Storybook · all subjects

ai/best-practices

5 notes, read out of this brain and free to use. Each one was extracted from a source and is re-checked against its exam.

Stories demonstrate one concept for AI agents

Stories should demonstrate one concept or use case and be descriptive about the 'why' behind the story, not just the 'what'. This helps AI agents understand when and why to use certain components or patterns.

Balance context in manifest for agent performance

Provide enough context so agents can use components effectively, but not so much that the manifest becomes overwhelming. Too little context prevents effective component use; too much irrelevant information leads to worse agent performance.

Critical instruction for agent component properties

Agents must never hallucinate component properties. Before using any property on a component, the agent MUST use MCP tools to verify the property is actually documented. Only use properties that are explicitly documented or shown in example stories.

Storybook AI capabilities preview status

Storybook's AI capabilities are currently in preview and only supported for React projects. The API may change in future releases.

Agent instructions for MCP server

Adjust your AGENTS.md (or CLAUDE.md for Claude) to guide your agent to use your MCP server. Include instructions to use docs-list and docs-show tools to access component documentation, verify properties before using them, use get-storybook-story-instructions tool for story creation conventions, and run test-run to check work.

Give your agent this brain