improvement(copilot): update copilot to match copilot repo

[Sg312]

Summary

Reverts complex streaming logics, handles proper thinking text scroll, adds gpt 5.2 and codex

Type of Change

• Bug fix
• New feature

Testing

Manual

Checklist

• Code follows project style guidelines
• Self-reviewed my changes
• Tests added/updated and passing
• No new warnings introduced
• I confirm that I have read and agree to the terms outlined in the Contributor License Agreement (CLA)

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

1 Skipped Deployment

Project	Deployment	Review	Updated (UTC)
docs	Skipped		Jan 15, 2026 4:48am

[greptile-apps]

Greptile Summary

This PR syncs copilot implementation with the standalone copilot repository by adding GPT 5.2 model variants, simplifying streaming text animation logic, and fixing UI issues.

Key changes:

• Added support for three new GPT 5.2 model variants (gpt-5.2, gpt-5.2-codex, gpt-5.2-pro) across API routes, type definitions, and UI components
• Simplified text streaming animation by removing complex adaptive delay algorithm in favor of fixed 3ms character delay
• Fixed page load flash by initializing displayed content based on streaming state in thinking and streaming components
• Added clearStreamingFlags function in store to prevent messages loaded from database from appearing as actively streaming
• Removed MAX badge display logic from model selector
• Hidden internal tool_search_tool_regex tool from UI rendering

Confidence Score: 5/5

• Safe to merge - well-structured changes with clear improvements to UX and no breaking changes
• All changes are straightforward and improve the codebase: new model support is consistently added across all necessary files, streaming simplification reduces complexity while maintaining functionality, and bug fixes address real UX issues without introducing risks
• No files require special attention

Important Files Changed

Filename	Overview
apps/sim/app/api/copilot/user-models/route.ts	Added GPT 5.2 models and updated default enabled models configuration, disabling 5.1 variants
apps/sim/app/workspace/[workspaceId]/w/[workflowId]/components/panel/components/copilot/components/copilot-message/components/smooth-streaming.tsx	Simplified streaming logic by removing adaptive delay algorithm, using fixed delay instead
apps/sim/stores/panel/copilot/store.ts	Added clearStreamingFlags function to prevent loaded messages from appearing as streaming

Sequence Diagram

sequenceDiagram
    participant User
    participant ModelSelector as Model Selector UI
    participant CopilotStore as Copilot Store
    participant ChatAPI as /api/copilot/chat
    participant UserModelsAPI as /api/copilot/user-models
    participant SmoothStreaming as Smooth Streaming Component
    participant ThinkingBlock as Thinking Block Component

    Note over User,ThinkingBlock: Model Selection & Configuration
    User->>UserModelsAPI: GET /api/copilot/user-models
    UserModelsAPI->>UserModelsAPI: Merge DEFAULT_ENABLED_MODELS<br/>(includes gpt-5.2 variants)
    UserModelsAPI-->>ModelSelector: Return enabled models
    ModelSelector->>ModelSelector: Render GPT 5.2 Codex, GPT 5.2 Pro options
    User->>ModelSelector: Select GPT 5.2 model

    Note over User,ThinkingBlock: Message Streaming Flow
    User->>CopilotStore: Send message with selected model
    CopilotStore->>ChatAPI: POST with model (gpt-5.2-*)
    ChatAPI->>ChatAPI: Validate model in ChatMessageSchema<br/>(includes new gpt-5.2 variants)
    ChatAPI-->>CopilotStore: Stream SSE events (thinking, content, tool_call)

    Note over SmoothStreaming,ThinkingBlock: UI Rendering with Simplified Streaming
    CopilotStore->>ThinkingBlock: Update thinking content (isStreaming=true)
    ThinkingBlock->>ThinkingBlock: Initialize displayedContent based on isStreaming
    ThinkingBlock->>ThinkingBlock: Animate with 0.5ms delay, 3 chars/frame
    CopilotStore->>SmoothStreaming: Update message content (isStreaming=true)
    SmoothStreaming->>SmoothStreaming: Initialize displayedContent based on isStreaming
    SmoothStreaming->>SmoothStreaming: Animate with fixed 3ms CHARACTER_DELAY

    Note over User,ThinkingBlock: Loading Persisted Messages
    User->>CopilotStore: Load chat from DB
    CopilotStore->>CopilotStore: normalizeMessagesForUI()<br/>clearStreamingFlags() on tool calls
    CopilotStore->>ThinkingBlock: Render thinking (isStreaming=false)
    ThinkingBlock->>ThinkingBlock: Initialize with full content (no flash)
    CopilotStore->>SmoothStreaming: Render content (isStreaming=false)
    SmoothStreaming->>SmoothStreaming: Initialize with full content (no flash)

Loading