Grok Build Plugin Marketplace Launches With Six Integrations; xAI SDK Adds Service Tiers and Image File Storage

Grok Build Plugin Marketplace launches with MongoDB, Vercel, Sentry, Chrome DevTools, Cloudflare, and Superpowers

• Launched June 11, the Grok Build Plugin Marketplace opens with six integrations covering database management (MongoDB), deployment (Vercel), error tracking (Sentry), browser debugging (Chrome DevTools), edge networking (Cloudflare), and AI workflow orchestration (Superpowers).
• The marketplace lets Grok Build agents call these tools natively within a session, reducing the boilerplate needed to connect Grok to external services.

Editor’s read: The six-plugin launch covers the most common webapp development surface area — Vercel plus Cloudflare plus Sentry is the standard deployment stack for solo builders. This makes Grok Build viable for production-grade work without custom MCP servers.

xAI Python SDK v1.17.0: service_tier parameter for priority-processing requests

• Released June 12, v1.17.0 adds a service_tier field to API requests, allowing callers with eligible plans to route requests to priority-processing infrastructure.
• This is useful for latency-sensitive pipelines that need predictable response times even during peak load.

Editor’s read: If response time variance is painful in your Grok-powered agents, service_tier is the lever to pull — check your plan tier for eligibility first.

xAI Python SDK v1.16.0: Imagine file storage, file-ID inputs, grok-imagine model aliases

• Released June 10, v1.16.0 adds persistent file storage for Imagine-generated images, the ability to pass a file ID as input to subsequent image generation calls (enabling multi-turn image editing workflows), and new grok-imagine model aliases for cleaner model targeting in code.

Editor’s read: File-ID inputs finally enable multi-step image workflows — generate a base image, pass the file ID to edit it, refine iteratively. This is how Imagine becomes production-useful rather than a one-shot demo.

AI dev tool power rankings June 2026: OpenCode overtakes Cursor, Claude Opus 4.7 leads models

• Published June 8, LogRocket's monthly rankings show shifts in the competitive landscape, with OpenCode now leading over Cursor and Claude Opus 4.7 as the top coding model.
• Grok Build does not yet appear prominently in these rankings; the Plugin Marketplace launch may shift that picture in next month's edition.

Hackaday: Revisiting AI Coding Assistants — You're Holding It Wrong

• Published June 8, a re-evaluation of AI coding tool workflows arguing that most developers get the best results by treating AI as a fast-drafting collaborator requiring verification rather than an autonomous agent.
• The workflow insight applies directly to how you structure Grok Build sessions with the new Plugin Marketplace integrations.