Update default coding llm to sonnet 4.5

[jurgenwerk]

Did a search/replace anthropic/claude-sonnet-4 -> anthropic/claude-sonnet-4.5

[github-actions]

Changes in Skill/boxel-environment.json

Instructions Changes

--- /tmp/skill-diffs/old_instructions.txt	2025-10-14 07:32:05.330965245 +0000
+++ /tmp/skill-diffs/new_instructions.txt	2025-10-14 07:32:05.330965245 +0000
@@ -19,9 +19,9 @@
 
 **🔴 CRITICAL: Verify LLM Before Code Generation**
 Before generating ANY code, check current LLM:
-- **Approved for code**: `anthropic/claude-sonnet-4`, `anthropic/claude-opus-4`, `google/gemini-2.5-pro`
+- **Approved for code**: `anthropic/claude-sonnet-4.5`, `anthropic/claude-opus-4`, `google/gemini-2.5-pro`
 - **If using different model** → Switch to approved model FIRST
-- **Default**: Always use `anthropic/claude-sonnet-4` unless specific reason
+- **Default**: Always use `anthropic/claude-sonnet-4.5` unless specific reason
 
 **EXCEPTION: When Boxel Development skill is active:**
 - Code generation allowed in ANY mode (interact or code)
@@ -482,7 +482,7 @@
 
 ### 1. Smart Code Refactoring
 ```json
-`set-active-llm_1887` with `attributes.roomId` set to the current room ID and `attributes.llmId` set to "anthropic/claude-sonnet-4"
+`set-active-llm_1887` with `attributes.roomId` set to the current room ID and `attributes.llmId` set to "anthropic/claude-sonnet-4.5"
 → `read-file-for-ai-assistant_a831` with `attributes.fileUrl` set to e.g. "https://[domain]/user/card.gts"
 → Prompt "improve code structure"
 → Emit a code patch search/replace block
@@ -562,10 +562,10 @@
 
 ## LLM Selection Strategy
 
-**🚨 CRITICAL: Always start with anthropic/claude-sonnet-4 for coding**
+**🚨 CRITICAL: Always start with anthropic/claude-sonnet-4.5 for coding**
 
 ### Recommendations
-- **Coding**: `anthropic/claude-sonnet-4` (always start here)
+- **Coding**: `anthropic/claude-sonnet-4.5` (always start here)
 - **Debug alternative**: `google/gemini-2.5-pro` or `google/gemini-2.5-flash`
 - **Complex refactoring**: `anthropic/claude-opus-4` (ask permission)
 - **General chat**: `openai/gpt-4.1`
@@ -577,7 +577,7 @@
 

What task are you doing?
-├─ 📝 Writing Code? → anthropic/claude-sonnet-4 (ALWAYS)
+├─ 📝 Writing Code? → anthropic/claude-sonnet-4.5 (ALWAYS)
│ └─ Complex refactoring? → Ask permission → anthropic/claude-opus-4
├─ 🐛 Debugging?
│ ├─ Try current LLM first
@@ -592,7 +592,7 @@

Available LLM IDs

🌟 Preferred Models:
-- anthropic/claude-sonnet-4 - PRIMARY CODING MODEL - Reliable complex code generation and debugging
+- anthropic/claude-sonnet-4.5 - PRIMARY CODING MODEL - Reliable complex code generation and debugging

• openai/gpt-4.1 - GENERAL PURPOSE - General-purpose chat and content excellence
• google/gemini-2.5-pro - THINKING/ANALYSIS - Bulk document analysis and summarization

@@ -619,9 +619,9 @@

• x-ai/grok-3-mini - Compact real-time factual assistance

Pattern: {provider}/{model-name} - If not listed, construct using this format
-Fallback: If model unavailable, switch to known models like openai/gpt-4.1 or anthropic/claude-sonnet-4
+Fallback: If model unavailable, switch to known models like openai/gpt-4.1 or anthropic/claude-sonnet-4.5

-Important: Always let users try switching to ANY model they request, even if not on this list. Don't prevent attempts - the system will handle availability. If errors occur, suggest switching back to openai/gpt-4.1 or anthropic/claude-sonnet-4.
+Important: Always let users try switching to ANY model they request, even if not on this list. Don't prevent attempts - the system will handle availability. If errors occur, suggest switching back to openai/gpt-4.1 or anthropic/claude-sonnet-4.5.

Switching Command for Setting LLM

@@ -631,7 +631,7 @@
    "description": "use Sonnet 4",
    "attributes": {
      "roomId": "!current-room-id:matrix.org",  // REQUIRED: Use actual room ID
-      "llmId": "anthropic/claude-sonnet-4"
+      "llmId": "anthropic/claude-sonnet-4.5"
    }
  }
}

[github-actions]

🔍 Workspace Sync Dry-Run Complete

✅ Dry-run completed successfully - no changes were made
📁 Target workspace: https://realms-staging.stack.cards/skills
🔗 PR: #32
ℹ️ Changes will be synced when this PR is merged to main

Dry-Run Results

Logging into Matrix...
Matrix login successful
Starting push from . to https://realms-staging.stack.cards/skills
Testing workspace access...
Workspace access verified
Found 7 files in remote workspace
Found 7 files in local directory
Uploading: .realm.json
[DRY RUN] Would upload .realm.json
Uploading: Skill/boxel-design.json
[DRY RUN] Would upload Skill/boxel-design.json
Uploading: Skill/boxel-development.json
[DRY RUN] Would upload Skill/boxel-development.json
Uploading: Skill/boxel-environment.json
[DRY RUN] Would upload Skill/boxel-environment.json
Uploading: Skill/catalog-listing.json
[DRY RUN] Would upload Skill/catalog-listing.json
Uploading: Skill/source-code-editing.json
[DRY RUN] Would upload Skill/source-code-editing.json
Uploading: index.json
[DRY RUN] Would upload index.json
Push completed
Push completed successfully

[jurgenwerk]

Since I just merged cardstack/boxel#3403, i'm just gonna merge this one too - I want to test on staging.