update links

blogs/2023/11/13/vscode-copilot-smarter.md

@@ -84,7 +84,7 @@ Once the intent is clear, the `@workspace` agent has a much better chance of add
 
 ## Extensibility
 
-"VS Code is just a shell, you need extensions to make it shine!" – the usual meeting anthem of Microsoft teams working on VS Code extensions, proudly flaunting their role in VS Code's success. We, as the VS Code core team, totally agree with them – VS Code would not be the product it is today without the rich extension ecosystem! AI is no different, and while the Core AI experience lights up with Copilot, our vision is that all the extensions from our ecosystem can participate and allow the LLM models to have the best context and grounding possible. Today we lay the foundation for this vision by adding the [agent API](https://github.com/microsoft/vscode/blob/main/src/vscode-dts/vscode.proposed.chatAgents2.d.ts#L8) in [a proposed state](https://code.visualstudio.com/api/advanced-topics/using-proposed-api).
+"VS Code is just a shell, you need extensions to make it shine!" – the usual meeting anthem of Microsoft teams working on VS Code extensions, proudly flaunting their role in VS Code's success. We, as the VS Code core team, totally agree with them – VS Code would not be the product it is today without the rich extension ecosystem! AI is no different, and while the Core AI experience lights up with Copilot, our vision is that all the extensions from our ecosystem can participate and allow the LLM models to have the best context and grounding possible. Today we lay the foundation for this vision by adding the [agent API](https://github.com/microsoft/vscode/blob/main/src/vscode-dts/vscode.proposed.chatParticipant.d.ts) in [a proposed state](https://code.visualstudio.com/api/advanced-topics/using-proposed-api).
 
 The agent API allows extensions to contribute agents that can answer specific questions by the user. Both the `@workspace` and `@vscode` agents are implemented using this API. With agents, users can bring rich and up-to-date information from their inner and outer loop tools into AI conversations while staying in the editor flow. Agents are like experts for an area, and when a user explicitly mentions an @agent in their prompt, that prompt is forwarded to the extension that contributed that specific agent.
 
@@ -102,9 +102,9 @@ And because an agent has the current context, it can continue the discussion:
 
 ![@1es agent providing more details about using CFS](1es-agent-second.png)
 
-We are also adding an API that allows extensions to [get access to the LLM](https://github.com/microsoft/vscode/blob/main/src/vscode-dts/vscode.proposed.chatRequestAccess.d.ts) and can choose to use the LLM to process and answer the user query. Today this API is limited to those extensions that implement an agent. The agent API passes the exact user prompts to contributed agents, and with the LLM access - agents can conveniently transition those language prompts into specific backend API calls. We will handle the usage of this API with care and transparency so that users are aware how many requests and tokens have been used by an agent.
+We are also adding an API that allows extensions to get access to the LLM and can choose to use the LLM to process and answer the user query. Today this API is limited to those extensions that implement an agent. The agent API passes the exact user prompts to contributed agents, and with the LLM access - agents can conveniently transition those language prompts into specific backend API calls. We will handle the usage of this API with care and transparency so that users are aware how many requests and tokens have been used by an agent.
 
-The agents API is still [proposed](https://code.visualstudio.com/api/advanced-topics/using-proposed-api), and we are looking for feedback on how to improve it, with the goal of finalizing the API in the near future. You can already try out things today, and the best way to start is from our [agent extensibility sample](https://github.com/microsoft/vscode-extension-samples/tree/main/chat-agent-sample). We can't wait to see the AI-driven innovations for developers that you create.
+The agents API is still [proposed](https://code.visualstudio.com/api/advanced-topics/using-proposed-api), and we are looking for feedback on how to improve it, with the goal of finalizing the API in the near future. You can already try out things today, and the best way to start is from our [agent extensibility sample](https://github.com/microsoft/vscode-extension-samples/tree/main/chat-sample). We can't wait to see the AI-driven innovations for developers that you create.
 
 ## Convenience