Add gpt_5_5 and claude_opus_4_7 to InvokeLLMParams model union

[hadashirt]

Summary

• Adds gpt_5_5 and claude_opus_4_7 to the model union on InvokeLLMParams in integrations.types.ts

What drifted

apper's RuntimeModel enum has these two models but the SDK type was missing them:

// Before
model?: 'gpt_5_mini' | 'gemini_3_flash' | 'gpt_5' | 'gpt_5_4' | 'gemini_3_1_pro' | 'claude_sonnet_4_6' | 'claude_opus_4_6';

// After
model?: 'gpt_5_mini' | 'gemini_3_flash' | 'gpt_5' | 'gpt_5_4' | 'gpt_5_5' | 'gemini_3_1_pro' | 'claude_sonnet_4_6' | 'claude_opus_4_6' | 'claude_opus_4_7';

Companion docs PR: base44-dev/mintlify-docs (llm-models-fix branch)

Test plan

• Verify InvokeLLMParams.model includes gpt_5_5 and claude_opus_4_7

🤖 Generated with Claude Code

[claude]

Claude finished @hadashirt's task —— View job

---

Code Review for PR #163

I've reviewed the changes to add gpt_5_5 and claude_opus_4_7 to the InvokeLLMParams model union in src/modules/integrations.types.ts.

✅ Code Quality and Best Practices

The code changes follow the existing patterns correctly:

• The new model values are added to the existing union type maintaining alphabetical order of model families
• Documentation comment above the type is properly maintained
• TypeScript type safety is preserved

✅ No Potential Bugs or Issues

The changes are straightforward type additions with no runtime impact. The implementation correctly:

• Extends the existing string literal union type
• Maintains backward compatibility (existing code continues to work)
• Does not introduce any breaking changes

✅ Performance Considerations

No performance concerns - this is purely a TypeScript type definition change with no runtime overhead.

✅ Security Concerns

No security issues identified. The changes only affect type definitions and don't introduce any new execution paths or data handling.

⚠️ Test Coverage Recommendation

While the existing test file (tests/unit/integrations.test.ts) covers the InvokeLLM functionality, I recommend adding test cases for the new model types to ensure they're properly accepted by the API:

test('InvokeLLM should accept gpt_5_5 model', async () => {
  const params = {
    prompt: 'Test prompt',
    model: 'gpt_5_5' as const,
  };
  
  scope
    .post(`/api/apps/${appId}/integration-endpoints/Core/InvokeLLM`, params)
    .reply(200, 'Response');
  
  const result = await base44.integrations.Core.InvokeLLM(params);
  expect(result).toBe('Response');
});

test('InvokeLLM should accept claude_opus_4_7 model', async () => {
  const params = {
    prompt: 'Test prompt',
    model: 'claude_opus_4_7' as const,
  };
  
  scope
    .post(`/api/apps/${appId}/integration-endpoints/Core/InvokeLLM`, params)
    .reply(200, 'Response');
  
  const result = await base44.integrations.Core.InvokeLLM(params);
  expect(result).toBe('Response');
});

Summary

The PR looks good and ready to merge. The changes are minimal, focused, and follow the existing code patterns. The only suggestion is to add test coverage for the new model types to ensure they work as expected with the API.

[github-actions]

🚀 Package Preview Available!

---

Install this PR's preview build with npm:

npm i @base44-preview/sdk@0.8.26-pr.163.555a7df

Prefer not to change any import paths? Install using npm alias so your code still imports @base44/sdk:

npm i "@base44/sdk@npm:@base44-preview/sdk@0.8.26-pr.163.555a7df"

Or add it to your package.json dependencies:

{
  "dependencies": {
    "@base44/sdk": "npm:@base44-preview/sdk@0.8.26-pr.163.555a7df"
  }
}

• 📦 Preview Package: @base44-preview/sdk@0.8.26-pr.163.555a7df
• 🔗 View this commit on GitHub

---

_{Preview published to npm registry — try new features instantly!}