src/llm-tool.spec.ts (4)

107-141: Good addition of a complex structured tool test case

This test case effectively validates the integration between the GameCharacterV2 class and OpenAI's Chat Completions API. It tests the conversion of a complex class structure to an OpenAI tool and verifies that the response properly follows the expected schema.

However, I noticed that line 137 duplicates the assertion from line 127 (expect(data3.location).toBeDefined()).

Remove the duplicate assertion:

    expect(data3.status).toBeDefined();
-   expect(data3.location).toBeDefined();
    expect(data3.titles[0]).toBeDefined();

---

108-109: Consider enhancing the user prompt for better test reliability

The current user message is quite minimal. For a complex structure like GameCharacterV2, providing more specific guidance in the prompt could lead to more consistent test results.

Consider expanding the prompt to include more specific guidance about what kind of game character to create:

-   const userMessage = `You are a helpful AI assistant. Based on the following JSON schema for a game character, please generate a mock response that follows the schema exactly. Make the data realistic and consistent with a game character context.
-       `;
+   const userMessage = `You are a helpful AI assistant. Based on the following JSON schema for a game character, please generate a mock response that follows the schema exactly. Create a character for an RPG game with appropriate levels, stats, location, bank accounts, titles, and status. Make the data realistic and consistent with a game character context.`;

---

124-126: Consider adding error handling for tool calls parsing

The current implementation assumes that the tool call will always be successful and present in the response.

Add error handling to make the test more robust:

-   const data3: GameCharacterV2 = JSON.parse(
-     completion.choices[0].message?.tool_calls[0].function.arguments,
-   );
+   expect(completion.choices[0].message?.tool_calls?.length).toBeGreaterThan(0);
+   const toolCall = completion.choices[0].message?.tool_calls?.[0];
+   expect(toolCall).toBeDefined();
+   expect(toolCall.function.name).toBe('game_character_v2');
+   const data3: GameCharacterV2 = JSON.parse(toolCall.function.arguments);

---

127-140: Consider adding stronger schema validation

The current test only checks if properties are defined but doesn't validate their types or structure in depth.

Consider adding more comprehensive validation for at least a few of the properties:

    expect(data3.location).toBeDefined();
    expect(data3.location.city).toBeDefined();
    expect(data3.location.country).toBeDefined();
+   expect(typeof data3.location.city).toBe('string');
+   expect(typeof data3.location.country).toBe('string');
    expect(data3.banks[0].account).toBeDefined();
    expect(data3.banks[0].bankName).toBeDefined();
+   expect(typeof data3.banks[0].account).toBe('string');
+   expect(typeof data3.banks[0].bankName).toBe('string');
    expect(data3.level).toBeDefined();
+   expect(typeof data3.level).toBe('number');
    expect(data3.name).toBeDefined();
    expect(data3.rank).toBeDefined();
    expect(data3.status).toBeDefined();
    expect(data3.titles[0]).toBeDefined();
    expect(data3.scores[0]).toBeGreaterThan(0);
    expect(data3.availableStatuses[0]).toBeDefined();
+   expect(Array.isArray(data3.titles)).toBe(true);
+   expect(Array.isArray(data3.scores)).toBe(true);
+   expect(Array.isArray(data3.availableStatuses)).toBe(true);

Configuration used: CodeRabbit UI
Review profile: CHILL
Plan: Pro