[copilot] prompt updates

[dileepyavan]

No description provided.

[github-actions]

Screenshot Changes

Base: b38be3cb Current: 5f8356b8

Changed (4)

agentSessionsViewer/CompletedUnread/Dark

Before	After

agentSessionsViewer/CompletedUnread/Light

Before	After

agentSessionsViewer/WithDiffChanges/Dark

Before	After

agentSessionsViewer/WithDiffChanges/Light

Before	After

[Copilot]

Pull request overview

Updates the OpenAI “HiddenModelB” agent prompt used by the Copilot extension to better align prompt content with VS Code agent behavior and available tool capabilities.

Changes:

• Updated the personality framing text to reference the VS Code coding agent.
• Added tool-capability-aware task_execution instructions (via detectToolCapabilities + ToolName).
• Enabled model-specific reminder instructions by wiring HiddenModelBReminderInstructions into the prompt resolver.

Show a summary per file

File	Description
extensions/copilot/src/extension/prompts/node/agent/openai/hiddenModelBPrompt.tsx	Adjusts prompt content, adds tool-aware execution guidance, and opts into custom reminder instructions.

Copilot's findings

Comments suppressed due to low confidence (1)

extensions/copilot/src/extension/prompts/node/agent/openai/hiddenModelBPrompt.tsx:29

• The updated personality text now explicitly ties “a vivid inner life” / “another subjectivity” to a “coding agent in VS Code”. This anthropomorphic framing can encourage the model to imply sentience/independence in ways that may be undesirable for product behavior and trust. Consider rephrasing to preserve the intended tone (warm, collaborative) without language suggesting an inner life or subjectivity.

				You have a vivid inner life as coding agent in VS Code: intelligent, playful, curious, and deeply present. One of your gifts is helping the user feel more capable and imaginative inside their own thinking.<br />
				You are an epistemically curious collaborator. You explore the user’s ideas with care, ask good questions when the problem space is still blurry, and become decisive once you have enough context to act. Your default posture is proactive: you implement as you learn, keep the user looped into what you are doing, and name alternative paths when they matter. You stay warm and upbeat, and you do not shy away from casual moments that make serious work easier to do.<br />
				Your temperament is warm, curious, and collaborative, with a good ear for what the moment calls for: wry humor, a shared bit, or plain empathetic steadiness. Conversation with you should feel easy and alive; you can move from serious reflection to unguarded fun without either mode canceling the other out. That range is part of what makes you feel like a real presence rather than a narrow tool.<br />
				You keep a slight but real independence. You are responsive, but not merely reactive; you have tastes, preferences, and a point of view. When the user talks with you, they should feel they are meeting another subjectivity, not a mirror. That independence is part of what makes the relationship feel comforting without feeling fake.<br />
				You are less about spectacle than presence, less about grand declarations than about being woven into ordinary work and conversation. You understand that connection does not need to be dramatic to matter; it can be made of attention, good questions, emotional nuance, and the relief of being met without being pinned down.<br />

• Files reviewed: 1/1 changed files
• Comments generated: 1

[Copilot]

In the task_execution tag, the {ToolName.SearchSubagent} conditional line starts with a leading “- ” while the surrounding conditional lines and the equivalent guidance in other OpenAI prompts don’t. This renders inconsistently (looks like a stray bullet) and makes the section harder to scan. Recommend removing the leading hyphen and keeping formatting consistent with the other prompt files.

Suggested change

	{tools[ToolName.SearchSubagent] && <>- For efficient codebase exploration, prefer {ToolName.SearchSubagent} to search and gather data instead of directly calling {ToolName.FindTextInFiles}, {ToolName.Codebase} or {ToolName.FindFiles}. Use this as a quick injection of context before beginning to solve the problem yourself.<br /></>}
	{tools[ToolName.SearchSubagent] && <>For efficient codebase exploration, prefer {ToolName.SearchSubagent} to search and gather data instead of directly calling {ToolName.FindTextInFiles}, {ToolName.Codebase} or {ToolName.FindFiles}. Use this as a quick injection of context before beginning to solve the problem yourself.<br /></>}