Microsoft’s latest update to Visual Studio Code, released August 26, features a Rubber Duck agent for a second model opinion as well as UX improvements to the Agents window. VS Code also now can show recent Copilot or Claude agent sessions created in other applications and allow users to continue them.
Microsoft’s latest update to Visual Studio Code, released August 26, features a Rubber Duck agent for a second model opinion as well as UX improvements to the Agents window. VS Code also now can show recent Copilot or Claude agent sessions created in other applications and allow users to continue them.
Downloads of Visual Studio Code are available for Mac, Windows, and Linux.
VS Code 1.135 features Rubber Duck, an experimental AI mode that debuted in GitHub Copilot CLI in April. The Rubber Duck agent is a built-in critic that provides a constructive second opinion on plans, code, and tests of the primary agent, using a different AI model from the one driving the session. Rubber Duck helps surface missed details or edge cases, according to Microsoft.
Also with VS Code 1.135, the agent host now lets users connect to the same agent session from multiple VS Code windows. The agent host runs agent harnesses in a dedicated process based on the Agent Host Protocol. Its Copilot agent is powered by the Copilot SDK, which aligns its behavior and functionality with the Copilot CLI, the standalone GitHub Copilot app, and other Copilot products, Microsoft said.
In addition, the Sessions list in VS Code now can show recent Copilot or Claude agent sessions created in other applications. Users can select a session from the Sessions list to view the conversation and continue it in VS Code with their Copilot subscription.
To give better insight into chat usage, Microsoft has redesigned the chat response footer. Hovering over the footer shows a per-model breakdown of the input, cached input, and output tokens used in the chat turn. Also, Agents window UX improvements allow developers to work with a streamlined side layout, simplified session controls, and session information that is easier to find, Microsoft said.
In context
- Topic: Inteligencia Artificial — IA aplicada al negocio: casos, límites, costos y gobierno.
- Source: InfoWorld
- Published: 27/08/2026
Continue reading at the original source →
Excerpt published automatically by the site radar. The full text belongs to its publisher and is linked above.
Why it matters
The conversation about artificial intelligence has become noisy, and that makes it harder to see what matters. What matters is rarely the model. It is which concrete process ends up cheaper, faster or more reliable, and who is accountable when the answer comes out wrong.
I always separate two things that get mixed up: the demo and the operation. A demo needs to work once with someone watching. An operation needs to work a thousand times with nobody watching, with odd cases, dirty data, at two in the morning. The gap between the two is where the budget goes.
What usually goes wrong
The point where it usually breaks is the handover to operations. The team that built the pilot knows how to read the results; the team receiving it does not. Without training and written criteria, people start accepting everything the model says or ignoring it entirely, and both are equally expensive.
What to watch
- Cost per query at real volume, not pilot volume: the variable that produces the most surprises.
- Where the data ends up and under what contract, especially when customer information is involved.
- Whether the organisation can change provider without rebuilding everything, which is the proof it did not get locked in.
How I read this entry
My practical advice is to fix on day one what happens when the model is wrong: who reviews, how often, and at what point it gets switched off. It sounds defensive, but it is exactly what lets you be aggressive later, because the risk stopped being an unknown.
This entry is an excerpt from the original source, selected by the site radar. The commentary above is the site's own and does not belong to the cited publisher.
Living through this in your own team?
Open the chat and tell me how you're handling it. I'm interested in comparing notes.
