Q63: model calls are priced and agent and department budgets limit real spend

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

tests/unit/domain-services.test.ts

@@ -24,7 +24,7 @@ describe("domain services", () => {
     const connection = createConnection();
 
     const rows = connection.db.prepare("select version from schema_migrations order by version asc").all() as Array<{ version: number }>;
-    expect(rows.map((row) => row.version)).toEqual([1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71]);
+    expect(rows.map((row) => row.version)).toEqual([1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72]);
     connection.close();
   });
 

tests/unit/spend.test.ts

@@ -0,0 +1,29 @@
+import { describe, expect, test } from "vitest";
+import { readTokenUsage, requestCostCents } from "../../src/main/services/spend";
+
+describe("token usage and cost", () => {
+  test("reads usage from every provider shape, cached and reasoning tokens included", () => {
+    expect(readTokenUsage(JSON.stringify({ usage: { input_tokens: 1000, output_tokens: 200, input_tokens_details: { cached_tokens: 400 }, output_tokens_details: { reasoning_tokens: 50 } } })))
+      .toEqual({ inputTokens: 1000, outputTokens: 200, cachedInputTokens: 400, reasoningTokens: 50 });
+    expect(readTokenUsage(JSON.stringify({ usage: { input_tokens: 600, output_tokens: 80, cache_read_input_tokens: 300, cache_creation_input_tokens: 100 } })))
+      .toEqual({ inputTokens: 1000, outputTokens: 80, cachedInputTokens: 300, reasoningTokens: 0 });
+    expect(readTokenUsage(JSON.stringify({ usage: { inputTokens: 500, outputTokens: 70, cacheReadInputTokens: 100 } })))
+      .toEqual({ inputTokens: 600, outputTokens: 70, cachedInputTokens: 100, reasoningTokens: 0 });
+    expect(readTokenUsage(JSON.stringify({ usageMetadata: { promptTokenCount: 900, candidatesTokenCount: 60, thoughtsTokenCount: 40, cachedContentTokenCount: 200 } })))
+      .toEqual({ inputTokens: 900, outputTokens: 100, cachedInputTokens: 200, reasoningTokens: 40 });
+    const chatStream = ["{\"choices\":[{\"delta\":{\"content\":\"hi\"}}]}", "{\"choices\":[],\"usage\":{\"prompt_tokens\":120,\"completion_tokens\":30,\"prompt_tokens_details\":{\"cached_tokens\":20}}}", "[DONE]"];
+    expect(readTokenUsage(JSON.stringify(chatStream))).toEqual({ inputTokens: 120, outputTokens: 30, cachedInputTokens: 20, reasoningTokens: 0 });
+    const responsesStream = ["{\"type\":\"response.created\"}", "{\"type\":\"response.completed\",\"response\":{\"usage\":{\"input_tokens\":10,\"output_tokens\":5}}}"];
+    expect(readTokenUsage(JSON.stringify(responsesStream))).toEqual({ inputTokens: 10, outputTokens: 5, cachedInputTokens: 0, reasoningTokens: 0 });
+    const moonshotStream = ["{\"choices\":[{\"index\":0,\"delta\":{},\"finish_reason\":\"stop\",\"usage\":{\"prompt_tokens\":40,\"completion_tokens\":9}}]}"];
+    expect(readTokenUsage(JSON.stringify(moonshotStream))).toEqual({ inputTokens: 40, outputTokens: 9, cachedInputTokens: 0, reasoningTokens: 0 });
+    expect(readTokenUsage(JSON.stringify({ output: [] }))).toBeNull();
+  });
+
+  test("cost prices fresh input, cached input and output separately; no prices, no cost", () => {
+    const usage = { inputTokens: 1_000_000, outputTokens: 500_000, cachedInputTokens: 400_000, reasoningTokens: 0 };
+    expect(requestCostCents(usage, { inputCentsPerMtok: 250, cachedInputCentsPerMtok: 125, outputCentsPerMtok: 1000 })).toBe(150 + 50 + 500);
+    expect(requestCostCents(usage, { inputCentsPerMtok: 250, cachedInputCentsPerMtok: null, outputCentsPerMtok: 1000 })).toBe(250 + 500);
+    expect(requestCostCents(usage, { inputCentsPerMtok: null, cachedInputCentsPerMtok: null, outputCentsPerMtok: 1000 })).toBeNull();
+  });
+});