AI News for Vibe Coders — Daily: June 24, 2026

Vibe Code Academy daily AI news cover for June 24, 2026

Welcome to the daily AI news brief for vibe coders. Today is Wednesday, June 24, 2026. The theme is tooling consolidation: Claude lands inside JetBrains IDEs through GitHub Copilot, Copilot’s CLI ships a generally available terminal, and Google’s Imagen image models retire today. If you build on Shopify, Claude, GPT, Gemini, or Firebase, here is what moved and what to do about it.

TL;DR

Claude is now an agent provider in GitHub Copilot for JetBrains IDEs, in public preview.
GitHub Copilot CLI’s new terminal interface is generally available, with same-day bring-your-own-key support.
Google Imagen image models shut down today; migrate to the Gemini “Nano Banana” models.
Claude Fable 5 left standard plans; access now costs credits at $10/$50 per million tokens.
Gemini 3.5 Pro’s GA window runs through June 30, but it is still preview-only today.
Build of the day: wire a fallback image model before Imagen endpoints stop responding.

Claude joins GitHub Copilot for JetBrains IDEs `[STACK]`

What shipped. GitHub put Claude into public preview as an agent provider inside GitHub Copilot for JetBrains IDEs. After you install the Claude Code CLI and set its path under Settings > Tools > GitHub Copilot > Chat, Claude appears in the Copilot chat agent picker. The same release added organization and enterprise custom agents, a /models command, and a per-turn AI credits indicator.

Why it matters for vibe coders. You can drive Claude Code from inside PyCharm, WebStorm, or IntelliJ without switching windows, and teams can standardize agents centrally.

What to do today. Install the Claude Code CLI and try Claude from the agent picker on a low-risk task. Note it currently runs in bypass-permissions mode, so edits and tool calls are auto-approved. GitHub

GitHub Copilot CLI’s new terminal interface goes GA `[STACK]`

What shipped. GitHub Copilot CLI’s redesigned terminal interface moved from preview to generally available on June 23, and GitHub added bring-your-own-key (BYOK) support to its Copilot apps the same day. The CLI also lets you queue a follow-up message, steer the current request, or stop and send while a task is still running.

Why it matters for vibe coders. A stable terminal agent plus BYOK lets you script agentic workflows against your own model keys instead of metered defaults.

What to do today. Update the Copilot CLI and test the queue-and-steer controls on one multi-step task. GitHub

Google Imagen image models shut down today `[STACK]`

What shipped. Google’s Imagen image models are deprecated and shut down today, June 24, 2026. The Gemini image models, the “Nano Banana” family, are the named replacement. Calls to retired Imagen endpoints stop returning images after today.

Why it matters for vibe coders. If your storefront, product mockups, or design tooling generate images through Firebase or the Gemini API, a hard-coded Imagen model breaks today, and a blank product image is easy to ship unnoticed.

What to do today. Grep your code and Firebase configs for “imagen,” swap to a supported Gemini image model, and regenerate one asset end to end to confirm the path works. Firebase

Claude Fable 5 moves to paid usage credits `[STACK]`

What shipped. Claude Fable 5 left Pro, Max, Team, and seat-based Enterprise subscriptions on June 23, ending the complimentary window from its June 9 launch. Access now requires paid usage credits at $10 per million input tokens and $50 per million output tokens, roughly double Claude Opus 4.8. The claude-fable-5 API string still works.

Why it matters for vibe coders. A model you may have prototyped on for free is now the most expensive option in the lineup, billed per token.

What to do today. Audit any agent pointed at claude-fable-5 and route all but your highest-value calls to Opus 4.8. Build Fast with AI

Gemini 3.5 Pro’s GA window opens `[ECOSYSTEM]`

What shipped. Google’s Gemini 3.5 Pro is inside its general-availability window, with analyst tracking pointing to a launch between June 23 and June 30. As of today it remains a limited Vertex AI preview with no public GA announcement. Confirmed specs include a 2-million-token context window, the largest in any production frontier model, and a Deep Think reasoning mode gated to the Ultra tier.

Why it matters for vibe coders. Wiring production code to a preview endpoint that has not formally shipped is how a launch-day feature breaks.

What to do today. Keep Firebase and Gemini work pinned to a generally available model until the official changelog confirms GA. Build Fast with AI

Also worth noting

Shopify’s Spring ’26 Edition turned the Universal Commerce Protocol on by default for every store and made the AI Toolkit GA, so agents can read your catalog out of the box. Shopify
SpaceX’s $60 billion all-stock deal for Cursor’s parent, announced June 16, is a week old; reports describe an “Origin” repository platform aimed at GitHub. Build Fast with AI
OpenAI folded Astral’s uv and ruff into the Codex stack; both tools keep working, but governance now sits with a frontier lab and terms are undisclosed. Build Fast with AI

Build of the day

Spend two hours making one app survive a model outage. Wrap your image-generation call in a handler that catches a model-unavailable error, retries against a second provider, and logs which one answered. Set a supported Gemini image model as primary so today’s Imagen shutdown cannot leave a blank product image. Test it by pointing the primary model string at an unresolvable name and confirming the request still completes.

FAQ

How do I use Claude inside JetBrains IDEs?

Install the Claude Code CLI, then open Settings > Tools > GitHub Copilot > Chat and set the CLI path. Claude then appears in the Copilot chat agent picker in public preview. It currently runs in bypass-permissions mode, so file edits and tool calls are auto-approved; configurable permissions are coming in a future release. GitHub

What happens to my app if it calls Google Imagen after June 24, 2026?

Google’s Imagen image models shut down today, June 24, 2026, with the Gemini “Nano Banana” image models offered as the replacement. After the shutdown, calls to retired Imagen endpoints stop returning images, which can leave product images blank. Search your code and Firebase configs now, swap to a supported Gemini image model, and verify one asset end to end. Firebase

Does Claude Fable 5 still cost extra after June 23, 2026?

Yes. Fable 5 left Pro, Max, Team, and seat-based Enterprise subscriptions on June 23, 2026. Access now requires paid usage credits at $10 per million input tokens and $50 per million output tokens, about double Claude Opus 4.8. The claude-fable-5 string still works. Build Fast with AI

Can I build on Gemini 3.5 Pro in production today?

Not yet. Gemini 3.5 Pro is in its GA window, with analysts expecting a launch between June 23 and June 30, 2026, but today it remains a limited Vertex AI preview with no public GA announcement. Pin Firebase and Gemini work to a generally available model until GA is confirmed. Build Fast with AI

What did GitHub change in the Copilot CLI on June 23?

GitHub made the Copilot CLI’s redesigned terminal interface generally available and added bring-your-own-key (BYOK) support to its Copilot apps. The CLI can now queue a follow-up message, steer the current request, or stop and send while a task is still running. Update the CLI to use the new controls. GitHub

Should Shopify developers act on the Universal Commerce Protocol now?

Yes. Shopify’s Spring ’26 Edition enabled the Universal Commerce Protocol by default on every store and made the AI Toolkit generally available, so AI shopping agents can read your catalog and build carts without a custom build. Confirm your product data is structured and complete so agents represent your store accurately. Shopify

Sources

About the author

Robert McCullock builds AI-assisted commerce on Shopify, Claude, and Firebase at Design Delight Studio. See his background and project work in his professional portfolio.