MCP

Use MCP when a hosted, shell-less AI assistant needs workspace-aware context or schema inspection. Shell-capable coding agents should use the CLI; hosted MCP is a single HTTP/OAuth surface for Console-backed discovery tools: get-collection-schema, get-tenant-context, and check-feature-progress. The local stdio transport was retired. MCP has no collection CRUD or commerce tools; use the CLI for shell agents and the SDK for application runtimes. MCP does not replace human approval for scope, credentials, or launch readiness.

get-collection-schema may include active custom-field structure for customers and products under optional collection.customFields: machine keys, types, required markers, select storage values, and the customData storage path. It does not return operator labels or stored customer/product values, and it does not expand MCP permissions.

Surfaces

Agent Handoff

• Give the agent the workspace goal, enabled features, and allowed change scope.
• For ecommerce work, ask the agent to run check-feature-progress before handoff.
• Keep Secret Key handling in trusted local or server environments.
• Ask the agent to report verification commands and changed files.
• Review workspace changes before launch.

Stop Conditions

• credential owner is unknown
• agent scope is broader than the task
• plan-blocked features are requested
• launch readiness has unresolved blockers

Next Actions

• Confirm owners in Integrations & Keys.
• Use CLI when the client can run shell commands.
• Check implementation contracts in API.
• Finish with Launch Readiness.