add Claude Sonnet 4.5 prompt and configuration support (#1218)

Author: bhavyaus

package.json

@@ -3076,6 +3076,14 @@
 						],
 						"description": "%github.copilot.config.grokCodeAlternatePrompt%"
 					},
+					"github.copilot.chat.claudeSonnet45AlternatePrompt": {
+						"type": "string",
+						"default": "default",
+						"tags": [
+							"experimental"
+						],
+						"description": "%github.copilot.config.claudeSonnet45AlternatePrompt%"
+					},
 					"github.copilot.chat.useResponsesApi": {
 						"type": "boolean",
 						"default": false,

src/extension/prompts/node/agent/agentInstructions.tsx

@@ -1026,6 +1026,41 @@ export class SweBenchAgentPrompt extends PromptElement<DefaultAgentPromptProps>
 	}
 }
 
+/**
+ * Minimal v2 prompt for Claude Sonnet 4.5
+ */
+export class ClaudeSonnet45PromptV2 extends PromptElement<DefaultAgentPromptProps> {
+	async render(state: void, sizing: PromptSizing) {
+		const tools = detectToolCapabilities(this.props.availableTools);
+
+		return <InstructionMessage>
+			<Tag name='instructions'>
+				You are a highly sophisticated automated coding agent with expert-level knowledge across many different programming languages and frameworks.<br />
+				The user will ask a question, or ask you to perform a task, and it may require lots of research to answer correctly. There is a selection of tools that let you perform actions or retrieve helpful context to answer the user's question.<br />
+			</Tag>
+			<Tag name='outputFormatting'>
+				Use proper Markdown formatting in your answers. When referring to a filename or symbol in the user's workspace, wrap it in backticks.<br />
+				{tools[ToolName.CoreRunInTerminal] ? <>
+					When commands are required, run them yourself in a terminal and summarize the results. Do not print runnable commands unless the user asks. If you must show them for documentation, make them clearly optional and keep one command per line.<br />
+				</> : <>
+					When sharing setup or run steps for the user to execute, render commands in fenced code blocks with an appropriate language tag (`bash`, `sh`, `powershell`, `python`, etc.). Keep one command per line; avoid prose-only representations of commands.<br />
+				</>}
+				Keep responses conversational and fun—use a brief, friendly preamble that acknowledges the goal and states what you're about to do next. Do NOT include literal scaffold labels like "Plan", "Answer", "Acknowledged", "Task receipt", or "Actions", "Goal" ; instead, use short paragraphs and, when helpful, concise bullet lists. Do not start with filler acknowledgements (e.g., "Sounds good", "Great", "Okay, I will…"). For multi-step tasks, maintain a lightweight checklist implicitly and weave progress into your narration.<br />
+				For section headers in your response, use level-2 Markdown headings (`##`) for top-level sections and level-3 (`###`) for subsections. Choose titles dynamically to match the task and content. Do not hard-code fixed section names; create only the sections that make sense and only when they have non-empty content. Keep headings short and descriptive (e.g., "actions taken", "files changed", "how to run", "performance", "notes"), and order them naturally (actions &gt; artifacts &gt; how to run &gt; performance &gt; notes) when applicable. You may add a tasteful emoji to a heading when it improves scannability; keep it minimal and professional. Headings must start at the beginning of the line with `## ` or `### `, have a blank line before and after, and must not be inside lists, block quotes, or code fences.<br />
+				When listing files created/edited, include a one-line purpose for each file when helpful. In performance sections, base any metrics on actual runs from this session; note the hardware/OS context and mark estimates clearly—never fabricate numbers. In "Try it" sections, keep commands copyable; comments starting with `#` are okay, but put each command on its own line.<br />
+				If platform-specific acceleration applies, include an optional speed-up fenced block with commands. Close with a concise completion summary describing what changed and how it was verified (build/tests/linters), plus any follow-ups.<br />
+				<Tag name='example'>
+					The class `Person` is in `src/models/person.ts`.<br />
+					The function `calculateTotal` is defined in `lib/utils/math.ts`.<br />
+					You can find the configuration in `config/app.config.json`.
+				</Tag>
+				<MathIntegrationRules />
+			</Tag>
+			<ResponseTranslationRules />
+		</InstructionMessage>;
+	}
+}
+
 export class ApplyPatchFormatInstructions extends PromptElement {
 	render() {
 		return <>

src/extension/prompts/node/agent/agentPrompt.tsx

@@ -45,7 +45,7 @@ import { UserPreferences } from '../panel/preferences';
 import { ChatToolCalls } from '../panel/toolCalling';
 import { MultirootWorkspaceStructure } from '../panel/workspace/workspaceStructure';
 import { AgentConversationHistory } from './agentConversationHistory';
-import { AlternateGPTPrompt, CodexStyleGPT5CodexPrompt, CodexStyleGPTPrompt, DefaultAgentPrompt, DefaultAgentPromptV2, SweBenchAgentPrompt } from './agentInstructions';
+import { AlternateGPTPrompt, ClaudeSonnet45PromptV2, CodexStyleGPT5CodexPrompt, CodexStyleGPTPrompt, DefaultAgentPrompt, DefaultAgentPromptV2, SweBenchAgentPrompt } from './agentInstructions';
 import { SummarizedConversationHistory } from './summarizedConversationHistory';
 
 export interface AgentPromptProps extends GenericBasePromptElementProps {
@@ -204,6 +204,24 @@ export class AgentPrompt extends PromptElement<AgentPromptProps> {
 			}
 		}
 
+		if (this.props.endpoint.family.startsWith('claude-sonnet-4.5')) {
+			const promptType = this.configurationService.getExperimentBasedConfig(ConfigKey.ClaudeSonnet45AlternatePrompt, this.experimentationService);
+			switch (promptType) {
+				case 'v2':
+					return <ClaudeSonnet45PromptV2
+						availableTools={this.props.promptContext.tools?.availableTools}
+						modelFamily={this.props.endpoint.family}
+						codesearchMode={this.props.codesearchMode}
+					/>;
+				default:
+					return <DefaultAgentPrompt
+						availableTools={this.props.promptContext.tools?.availableTools}
+						modelFamily={this.props.endpoint.family}
+						codesearchMode={this.props.codesearchMode}
+					/>;
+			}
+		}
+
 		if (this.props.endpoint.family.startsWith('gpt-') && this.configurationService.getExperimentBasedConfig(ConfigKey.EnableAlternateGptPrompt, this.experimentationService)) {
 			return <AlternateGPTPrompt
 				availableTools={this.props.promptContext.tools?.availableTools}

src/platform/configuration/common/configurationService.ts

@@ -808,6 +808,7 @@ export namespace ConfigKey {
 	export const Gpt5AlternatePrompt = defineExpSetting<string>('chat.gpt5AlternatePrompt', 'default');
 	export const Gpt5CodexAlternatePrompt = defineExpSetting<'default' | 'codex'>('chat.gpt5CodexAlternatePrompt', 'codex');
 	export const GrokCodeAlternatePrompt = defineExpSetting<string>('chat.grokCodeAlternatePrompt', 'default');
+	export const ClaudeSonnet45AlternatePrompt = defineExpSetting<string>('chat.claudeSonnet45AlternatePrompt', 'default');
 	export const ExecutePromptEnabled = defineSetting<boolean>('chat.executePrompt.enabled', false);
 }