[{"_1":2,"_3":-5,"_4":-5},"loaderData",{"_5":6,"_7":8,"_9":10},"actionData","errors","root",{"_220":221,"_222":223},"routes/_app.x.$handle.$name",{"_101":100,"_32":45,"_109":87,"_27":110,"_13":14,"_111":-5,"_11":112,"_113":114,"_115":116,"_117":118,"_119":120,"_20":-5,"_121":122,"_123":124,"_125":25,"_126":127,"_128":25,"_129":-5,"_130":-5,"_131":132},"routes/_app.x.$handle.$name.branch.$branch.version.$version.diff",{"_11":12,"_13":14,"_15":16,"_17":-5,"_18":-5},"project",{"_43":44,"_32":45,"_38":46,"_47":-5,"_48":25,"_49":50,"_51":50,"_52":53,"_54":55,"_18":56,"_57":58,"_59":25,"_60":-5,"_61":-5,"_62":63,"_64":65,"_66":-5,"_67":-5,"_68":-5,"_69":-5,"_70":71,"_72":71,"_73":74,"_75":74,"_76":74,"_77":71,"_78":71,"_79":80,"_81":82,"_83":84,"_85":25},"version",17,"changes",[19],"aiChatInfo","userId",{"_20":21,"_22":23,"_24":25,"_26":25,"_27":28,"_29":28},"file",{"_30":31,"_32":28,"_33":34,"_35":37,"_38":39,"_40":14,"_41":42,"_24":25},"previous",{"_30":31,"_32":28,"_33":34,"_35":36},"deleted",false,"deletedHere","path","cost.test.ts","latestPath","fileId","b62843ba-efdc-11f0-b032-42dde27851f2","name","fileType","script","content","import { expect } from \"jsr:@std/expect@1.0.17\";\n\n// Mock data based on actual SDK output from logs\nconst mockSimpleRun = {\n state: {\n _currentAgent: { model: \"gpt-5\" },\n _modelResponses: [{\n usage: {\n inputTokens: 15,\n outputTokens: 7,\n totalTokens: 22,\n inputTokensDetails: [{ cachedTokens: 0 }],\n }\n }],\n _generatedItems: [\n { type: \"reasoning_item\", rawItem: { type: \"reasoning\" } },\n { type: \"message_output_item\", rawItem: { type: \"message\" } },\n ]\n }\n};\n\nconst mockWebSearchRun = {\n state: {\n _currentAgent: { model: \"gpt-5\" },\n _modelResponses: [{\n usage: {\n inputTokens: 69153,\n outputTokens: 3858,\n totalTokens: 73011,\n inputTokensDetails: [{ cachedTokens: 10000 }],\n }\n }],\n _generatedItems: [\n { type: \"reasoning_item\", rawItem: { type: \"reasoning\" } },\n { type: \"tool_call_item\", rawItem: { type: \"hosted_tool_call\", name: \"web_search_call\" } },\n { type: \"tool_call_item\", rawItem: { type: \"hosted_tool_call\", name: \"web_search_call\" } },\n { type: \"message_output_item\", rawItem: { type: \"message\" } },\n ]\n }\n};\n\nconst mockVersionedModelRun = {\n state: {\n _currentAgent: { model: \"gpt-5-2025-08-07\" },\n _modelResponses: [{\n usage: {\n inputTokens: 1000,\n outputTokens: 500,\n inputTokensDetails: [],\n }\n }],\n _generatedItems: []\n }\n};\n\nconst mockMultiResponseRun = {\n state: {\n _currentAgent: { model: \"gpt-4.1-mini\" },\n _modelResponses: [\n { usage: { inputTokens: 100, outputTokens: 50, inputTokensDetails: [{ cachedTokens: 20 }] } },\n { usage: { inputTokens: 200, outputTokens: 100, inputTokensDetails: [{ cachedTokens: 50 }] } },\n ],\n _generatedItems: []\n }\n};\n\n// Run tests\nconsole.log(\"Running cost2.ts tests...\\n\");\n\n// Test 1: Simple run\n{\n const result = estimateCost(mockSimpleRun);\n \n expect(result.model).toBe(\"gpt-5\");\n expect(result.tokens.input).toBe(15);\n expect(result.tokens.output).toBe(7);\n expect(result.tokens.cachedInput).toBe(0);\n expect(result.tokens.nonCachedInput).toBe(15);\n expect(Object.keys(result.toolCalls).length).toBe(0);\n expect(result.toolCostUSD).toBe(0);\n \n // Token cost: (15/1M * $2.50) + (7/1M * $20.00)\n const expectedCost = (15 / 1_000_000) * 2.50 + (7 / 1_000_000) * 20.0;\n expect(result.tokenCostUSD).toBeCloseTo(expectedCost, 10);\n expect(result.totalCostUSD).toBeCloseTo(expectedCost, 10);\n \n console.log(\"✓ Test 1: Simple run - PASSED\");\n}\n\n// Test 2: Web search with cached tokens\n{\n const result = estimateCost(mockWebSearchRun);\n \n expect(result.model).toBe(\"gpt-5\");\n expect(result.tokens.input).toBe(69153);\n expect(result.tokens.cachedInput).toBe(10000);\n expect(result.tokens.nonCachedInput).toBe(59153);\n expect(result.tokens.output).toBe(3858);\n \n expect(result.toolCalls[\"web_search\"]).toBe(2);\n expect(result.toolCostUSD).toBeCloseTo(0.02, 10); // 2 calls * $0.01\n \n // Token cost calculation\n const pricing = MODEL_PRICES[\"gpt-5\"];\n const expectedTokenCost = \n (59153 / 1_000_000) * pricing.input +\n (10000 / 1_000_000) * pricing.cachedInput +\n (3858 / 1_000_000) * pricing.output;\n \n expect(result.tokenCostUSD).toBeCloseTo(expectedTokenCost, 10);\n expect(result.totalCostUSD).toBeCloseTo(expectedTokenCost + 0.02, 10);\n \n console.log(\"✓ Test 2: Web search with cached tokens - PASSED\");\n}\n\n// Test 3: Versioned model name normalization\n{\n const result = estimateCost(mockVersionedModelRun);\n \n expect(result.model).toBe(\"gpt-5\");\n expect(result.tokens.input).toBe(1000);\n expect(result.tokens.output).toBe(500);\n expect(result.tokens.cachedInput).toBe(0);\n \n console.log(\"✓ Test 3: Versioned model name normalization - PASSED\");\n}\n\n// Test 4: Multiple model responses aggregation\n{\n const result = estimateCost(mockMultiResponseRun);\n \n expect(result.model).toBe(\"gpt-4.1-mini\");\n expect(result.tokens.input).toBe(300); // 100 + 200\n expect(result.tokens.output).toBe(150); // 50 + 100\n expect(result.tokens.cachedInput).toBe(70); // 20 + 50\n expect(result.tokens.nonCachedInput).toBe(230);\n \n console.log(\"✓ Test 4: Multiple model responses aggregation - PASSED\");\n}\n\n// Test 5: Custom model pricing\n{\n const customRun = {\n state: {\n _currentAgent: { model: \"my-custom-model\" },\n _modelResponses: [{\n usage: { inputTokens: 1000, outputTokens: 1000, inputTokensDetails: [] }\n }],\n _generatedItems: []\n }\n };\n \n const result = estimateCost(customRun, {\n modelPrices: {\n \"my-custom-model\": { input: 1.0, cachedInput: 0.1, output: 2.0 }\n }\n });\n \n expect(result.model).toBe(\"my-custom-model\");\n // (1000/1M * $1) + (1000/1M * $2) = $0.003\n expect(result.tokenCostUSD).toBeCloseTo(0.003, 10);\n \n console.log(\"✓ Test 5: Custom model pricing - PASSED\");\n}\n\n// Test 6: Unknown model throws error\n{\n const unknownRun = {\n state: {\n _currentAgent: { model: \"unknown-model\" },\n _modelResponses: [{ usage: { inputTokens: 100 } }],\n _generatedItems: []\n }\n };\n \n let threw = false;\n try {\n estimateCost(unknownRun);\n } catch (e) {\n threw = true;\n expect(e.message).toContain(\"No pricing for model\");\n }\n expect(threw).toBe(true);\n \n console.log(\"✓ Test 6: Unknown model throws error - PASSED\");\n}\n\n// Test 7: Tool name normalization\n{\n const toolRun = {\n state: {\n _currentAgent: { model: \"gpt-5\" },\n _modelResponses: [{ usage: { inputTokens: 100, outputTokens: 50, inputTokensDetails: [] } }],\n _generatedItems: [\n { type: \"tool_call_item\", rawItem: { name: \"web_search_call\" } },\n { type: \"tool_call_item\", rawItem: { name: \"file_search_call\" } },\n { type: \"tool_call_item\", rawItem: { type: \"code_interpreter\" } },\n { type: \"tool_call_item\", rawItem: { name: \"my_custom_tool_call\" } },\n ]\n }\n };\n \n const result = estimateCost(toolRun);\n \n expect(result.toolCalls[\"web_search\"]).toBe(1);\n expect(result.toolCalls[\"file_search\"]).toBe(1);\n expect(result.toolCalls[\"code_interpreter_session\"]).toBe(1);\n expect(result.toolCalls[\"my_custom_tool\"]).toBe(1);\n \n // Known tools: $0.01 + $0.0025 + $0.03 = $0.0425\n // Custom tool: $0 (not in price list)\n expect(result.toolCostUSD).toBeCloseTo(0.0425, 10);\n \n console.log(\"✓ Test 7: Tool name normalization - PASSED\");\n}\n\n// Test 8: Empty/malformed input handling\n{\n const emptyResult = estimateCost({}, { \n defaultModel: \"gpt-5\",\n modelPrices: MODEL_PRICES \n });\n \n expect(emptyResult.model).toBe(\"gpt-5\");\n expect(emptyResult.tokens.input).toBe(0);\n expect(emptyResult.tokens.output).toBe(0);\n expect(emptyResult.tokenCostUSD).toBe(0);\n \n console.log(\"✓ Test 8: Empty/malformed input handling - PASSED\");\n}\n\nconsole.log(\"\\n✅ All tests passed!\");\n\n// Print example output\nconsole.log(\"\\n--- Example Output ---\");\nconsole.log(\"Simple run cost:\", estimateCost(mockSimpleRun));\nconsole.log(\"\\nWeb search run cost:\", estimateCost(mockWebSearchRun));","import { expect } from \"jsr:@std/expect@1.0.17\";\nimport { estimateCost } from \"./cost.ts\";\n\n// Mock data based on actual SDK output from logs\nconst mockSimpleRun = {\n state: {\n _currentAgent: { model: \"gpt-5\" },\n _modelResponses: [{\n usage: {\n inputTokens: 15,\n outputTokens: 7,\n totalTokens: 22,\n inputTokensDetails: [{ cachedTokens: 0 }],\n }\n }],\n _generatedItems: [\n { type: \"reasoning_item\", rawItem: { type: \"reasoning\" } },\n { type: \"message_output_item\", rawItem: { type: \"message\" } },\n ]\n }\n};\n\nconst mockWebSearchRun = {\n state: {\n _currentAgent: { model: \"gpt-5\" },\n _modelResponses: [{\n usage: {\n inputTokens: 69153,\n outputTokens: 3858,\n totalTokens: 73011,\n inputTokensDetails: [{ cachedTokens: 10000 }],\n }\n }],\n _generatedItems: [\n { type: \"reasoning_item\", rawItem: { type: \"reasoning\" } },\n { type: \"tool_call_item\", rawItem: { type: \"hosted_tool_call\", name: \"web_search_call\" } },\n { type: \"tool_call_item\", rawItem: { type: \"hosted_tool_call\", name: \"web_search_call\" } },\n { type: \"message_output_item\", rawItem: { type: \"message\" } },\n ]\n }\n};\n\nconst mockVersionedModelRun = {\n state: {\n _currentAgent: { model: \"gpt-5-2025-08-07\" },\n _modelResponses: [{\n usage: {\n inputTokens: 1000,\n outputTokens: 500,\n inputTokensDetails: [],\n }\n }],\n _generatedItems: []\n }\n};\n\nconst mockMultiResponseRun = {\n state: {\n _currentAgent: { model: \"gpt-4.1-mini\" },\n _modelResponses: [\n { usage: { inputTokens: 100, outputTokens: 50, inputTokensDetails: [{ cachedTokens: 20 }] } },\n { usage: { inputTokens: 200, outputTokens: 100, inputTokensDetails: [{ cachedTokens: 50 }] } },\n ],\n _generatedItems: []\n }\n};\n\n// Run tests\nconsole.log(\"Running cost2.ts tests...\\n\");\n\n// Test 1: Simple run\n{\n const result = estimateCost(mockSimpleRun);\n \n expect(result.model).toBe(\"gpt-5\");\n expect(result.tokens.input).toBe(15);\n expect(result.tokens.output).toBe(7);\n expect(result.tokens.cachedInput).toBe(0);\n expect(result.tokens.nonCachedInput).toBe(15);\n expect(Object.keys(result.toolCalls).length).toBe(0);\n expect(result.toolCostUSD).toBe(0);\n \n // Token cost: (15/1M * $2.50) + (7/1M * $20.00)\n const expectedCost = (15 / 1_000_000) * 2.50 + (7 / 1_000_000) * 20.0;\n expect(result.tokenCostUSD).toBeCloseTo(expectedCost, 10);\n expect(result.totalCostUSD).toBeCloseTo(expectedCost, 10);\n \n console.log(\"✓ Test 1: Simple run - PASSED\");\n}\n\n// Test 2: Web search with cached tokens\n{\n const result = estimateCost(mockWebSearchRun);\n \n expect(result.model).toBe(\"gpt-5\");\n expect(result.tokens.input).toBe(69153);\n expect(result.tokens.cachedInput).toBe(10000);\n expect(result.tokens.nonCachedInput).toBe(59153);\n expect(result.tokens.output).toBe(3858);\n \n expect(result.toolCalls[\"web_search\"]).toBe(2);\n expect(result.toolCostUSD).toBeCloseTo(0.02, 10); // 2 calls * $0.01\n \n // Token cost calculation\n const pricing = MODEL_PRICES[\"gpt-5\"];\n const expectedTokenCost = \n (59153 / 1_000_000) * pricing.input +\n (10000 / 1_000_000) * pricing.cachedInput +\n (3858 / 1_000_000) * pricing.output;\n \n expect(result.tokenCostUSD).toBeCloseTo(expectedTokenCost, 10);\n expect(result.totalCostUSD).toBeCloseTo(expectedTokenCost + 0.02, 10);\n \n console.log(\"✓ Test 2: Web search with cached tokens - PASSED\");\n}\n\n// Test 3: Versioned model name normalization\n{\n const result = estimateCost(mockVersionedModelRun);\n \n expect(result.model).toBe(\"gpt-5\");\n expect(result.tokens.input).toBe(1000);\n expect(result.tokens.output).toBe(500);\n expect(result.tokens.cachedInput).toBe(0);\n \n console.log(\"✓ Test 3: Versioned model name normalization - PASSED\");\n}\n\n// Test 4: Multiple model responses aggregation\n{\n const result = estimateCost(mockMultiResponseRun);\n \n expect(result.model).toBe(\"gpt-4.1-mini\");\n expect(result.tokens.input).toBe(300); // 100 + 200\n expect(result.tokens.output).toBe(150); // 50 + 100\n expect(result.tokens.cachedInput).toBe(70); // 20 + 50\n expect(result.tokens.nonCachedInput).toBe(230);\n \n console.log(\"✓ Test 4: Multiple model responses aggregation - PASSED\");\n}\n\n// Test 5: Custom model pricing\n{\n const customRun = {\n state: {\n _currentAgent: { model: \"my-custom-model\" },\n _modelResponses: [{\n usage: { inputTokens: 1000, outputTokens: 1000, inputTokensDetails: [] }\n }],\n _generatedItems: []\n }\n };\n \n const result = estimateCost(customRun, {\n modelPrices: {\n \"my-custom-model\": { input: 1.0, cachedInput: 0.1, output: 2.0 }\n }\n });\n \n expect(result.model).toBe(\"my-custom-model\");\n // (1000/1M * $1) + (1000/1M * $2) = $0.003\n expect(result.tokenCostUSD).toBeCloseTo(0.003, 10);\n \n console.log(\"✓ Test 5: Custom model pricing - PASSED\");\n}\n\n// Test 6: Unknown model throws error\n{\n const unknownRun = {\n state: {\n _currentAgent: { model: \"unknown-model\" },\n _modelResponses: [{ usage: { inputTokens: 100 } }],\n _generatedItems: []\n }\n };\n \n let threw = false;\n try {\n estimateCost(unknownRun);\n } catch (e) {\n threw = true;\n expect(e.message).toContain(\"No pricing for model\");\n }\n expect(threw).toBe(true);\n \n console.log(\"✓ Test 6: Unknown model throws error - PASSED\");\n}\n\n// Test 7: Tool name normalization\n{\n const toolRun = {\n state: {\n _currentAgent: { model: \"gpt-5\" },\n _modelResponses: [{ usage: { inputTokens: 100, outputTokens: 50, inputTokensDetails: [] } }],\n _generatedItems: [\n { type: \"tool_call_item\", rawItem: { name: \"web_search_call\" } },\n { type: \"tool_call_item\", rawItem: { name: \"file_search_call\" } },\n { type: \"tool_call_item\", rawItem: { type: \"code_interpreter\" } },\n { type: \"tool_call_item\", rawItem: { name: \"my_custom_tool_call\" } },\n ]\n }\n };\n \n const result = estimateCost(toolRun);\n \n expect(result.toolCalls[\"web_search\"]).toBe(1);\n expect(result.toolCalls[\"file_search\"]).toBe(1);\n expect(result.toolCalls[\"code_interpreter_session\"]).toBe(1);\n expect(result.toolCalls[\"my_custom_tool\"]).toBe(1);\n \n // Known tools: $0.01 + $0.0025 + $0.03 = $0.0425\n // Custom tool: $0 (not in price list)\n expect(result.toolCostUSD).toBeCloseTo(0.0425, 10);\n \n console.log(\"✓ Test 7: Tool name normalization - PASSED\");\n}\n\n// Test 8: Empty/malformed input handling\n{\n const emptyResult = estimateCost({}, { \n defaultModel: \"gpt-5\",\n modelPrices: MODEL_PRICES \n });\n \n expect(emptyResult.model).toBe(\"gpt-5\");\n expect(emptyResult.tokens.input).toBe(0);\n expect(emptyResult.tokens.output).toBe(0);\n expect(emptyResult.tokenCostUSD).toBe(0);\n \n console.log(\"✓ Test 8: Empty/malformed input handling - PASSED\");\n}\n\nconsole.log(\"\\n✅ All tests passed!\");\n\n// Print example output\nconsole.log(\"\\n--- Example Output ---\");\nconsole.log(\"Simple run cost:\", estimateCost(mockSimpleRun));\nconsole.log(\"\\nWeb search run cost:\", estimateCost(mockWebSearchRun));","createdAt",["D",1768239189321],"branchVersion","branchId","62a92182-efd7-11f0-8671-42dde27851f2","id","62a5131c-efd7-11f0-8671-42dde27851f2","openai-agents",["D",1768236880010],"deletedAt","legacyEnvVars","privacy","public","httpPrivacy","description","Template to use the OpenAI Agents SDK","imageUrl","https://imagedelivery.net/iHX6Ovru0O7AjmyT5yZRoA/4808a810-6001-47ed-dac6-b18afdf23100/square512?exp=1789592399&sig=5750d4a52baa304531bd8e41d18640515f200338e7cb6e7c716f9d2187b79960","a0bf3b31-15a5-4d5c-880e-4b1e22c9bc18","pinnedFileId","019bb321-90d8-7149-add7-53d9b5c4e821","pinPreview","pinMigrated","screenshotVersion","likeCount",0,"orgId","e81e7335-6d4c-4764-a902-d0962bbfa9f7","tursoName","tursoHostname","tursoMeasuredAt","pinned","userPerm","r","valPerm","blobPerm","rw","emailPerm","sqlitePerm","projectPerm","telemetryPerm","user",{"_101":100,"_108":104},"org",{"_101":100,"_102":-5,"_103":104,"_18":56,"_105":106,"_107":25},"branch",{"_13":86,"_32":87,"_43":42,"_88":44,"_89":90,"_91":92,"_93":50,"_94":95,"_96":97,"_98":45,"_99":100},"isOwned",21,"main","projectId","forkedBranchId","52cd3bfc-c630-11f0-a182-42dde27851f2","forkedBranchVersion",50,"forkedProjectPrivacy","forkedProjectId","52ccf82c-c630-11f0-a182-42dde27851f2","forkedProjectOrgId","019aa807-52c2-709c-928b-2029bae9a558","projectName","userHandle","stevekrouse","handle","displayName","avatar","https://images.clerk.dev/uploaded/img_2PqHa2Gsy93xQrjh2w78Xu0cChW.jpeg","tier","pro","canRestrictVals","profileImageUrl","branchName","","adoptOrgId",{"_43":44,"_32":45,"_38":216,"_47":-5,"_48":25,"_49":50,"_51":50,"_52":53,"_54":55,"_18":56,"_57":58,"_59":25,"_60":-5,"_61":-5,"_62":63,"_64":65,"_66":-5,"_67":-5,"_68":-5,"_69":-5,"_70":71,"_72":71,"_73":74,"_75":74,"_76":74,"_77":71,"_78":71,"_79":217,"_81":218,"_83":219,"_85":25},"tree",[195,196,197,198,199],"files",[195,196,197,198,199],"counts",{"_185":63,"_186":181,"_187":63,"_188":189,"_190":191},"contents",[143,144,145,146,147],"upstreamProject",{"_43":95,"_32":45,"_52":53,"_54":140,"_49":50,"_79":141},"upstreamBranch",{"_43":90,"_38":133,"_134":135,"_32":87,"_88":95,"_13":136,"_89":-5,"_91":-5,"_137":138,"_47":-5,"_85":25,"_49":50,"_99":139,"_64":97,"_98":45},"emptyEnvVars","flags",[],"hasHttpFiles","grants","orgs","customDomains",{},["D",1763657129834],"updatedAt",["D",1775215982290],60,"lockfileKey","1773770259118-esm.town_v_templates_openai-agents@57-main","templates","https://imagedelivery.net/iHX6Ovru0O7AjmyT5yZRoA/8526e049-b3af-490b-bd8e-ef1febbd0600/square512?exp=1789592399&sig=fef467cf00002d3b095539f5f53c7ce2cd7d571fe291a2330da6e935783763dd",{"_101":139,"_108":142},"https://imagedelivery.net/iHX6Ovru0O7AjmyT5yZRoA/129cab32-3bd7-4c3e-f39b-f12900c76100/square512?exp=1789592399&sig=e3667930a878fd5db49e47cce78fe270a8ba33317e34775a1ada2dd4a7c49d8d",{"_43":178,"_32":179,"_30":180,"_150":-5,"_41":42,"_40":181,"_152":153,"_38":182,"_13":63,"_33":20,"_35":183,"_24":25,"_157":-5,"_27":184},{"_43":174,"_32":28,"_30":31,"_150":-5,"_41":42,"_40":14,"_152":153,"_38":175,"_13":176,"_33":34,"_35":37,"_24":25,"_157":-5,"_27":177},{"_43":166,"_32":167,"_30":168,"_150":-5,"_41":42,"_40":169,"_152":25,"_38":170,"_13":171,"_33":34,"_35":172,"_24":25,"_157":-5,"_27":173},{"_43":159,"_32":160,"_30":161,"_150":-5,"_41":42,"_40":162,"_152":25,"_38":163,"_13":63,"_33":34,"_35":164,"_24":25,"_157":-5,"_27":165},{"_43":148,"_32":149,"_30":58,"_150":-5,"_41":42,"_40":151,"_152":153,"_38":154,"_13":155,"_33":34,"_35":156,"_24":25,"_157":-5,"_27":158},"97299306-efd7-11f0-b8ed-42dde27851f2","main.ts","parentId",5,"latest",true,["D",1768236968140],4,"import { Agent, run, webSearchTool } from \"npm:@openai/agents@0.3.0\";\n\nconst instructions = `You research the web looking for news stories\nabout topical events happening within the last 24 hours.\nProvide a summary of these events as a bulleted list with links to sources.`;\n\nconst agent = new Agent({\n name: \"Agent\",\n // instructions,\n model: \"gpt-5\",\n // tools: [webSearchTool()],\n});\n\nconst result = await run(agent, \"This is a test. Reply 'hi'\");\nconsole.log(result);","label",[149],"b6edeb92-efdc-11f0-b34c-42dde27851f2","cost2.ts","b6edec32-efdc-11f0-b34c-42dde27851f2",14,["D",1768239168920],"/**\n * Cost estimation for @openai/agents SDK run results.\n * Simplified implementation targeting the actual SDK structure.\n */\n\n// ============================================================================\n// Types\n// ============================================================================\n\ntype ModelPricing = {\n input: number; // $ per 1M input tokens\n cachedInput: number; // $ per 1M cached input tokens\n output: number; // $ per 1M output tokens\n};\n\ntype CostBreakdown = {\n model: string;\n tokens: {\n input: number;\n cachedInput: number;\n nonCachedInput: number;\n output: number;\n };\n tokenCostUSD: number;\n toolCalls: Record;\n toolCostUSD: number;\n totalCostUSD: number;\n};\n\ntype EstimateOptions = {\n modelPrices?: Record;\n toolPrices?: Record;\n defaultModel?: string;\n};\n\n// ============================================================================\n// Default Pricing (Jan 2026 - verify at https://openai.com/api/pricing/)\n// ============================================================================\n\nexport const MODEL_PRICES: Record = {\n // GPT-5 series\n \"gpt-5\": { input: 2.50, cachedInput: 0.25, output: 20.0 },\n \"gpt-5-mini\": { input: 0.25, cachedInput: 0.025, output: 2.0 },\n\n // GPT-4.1 series\n \"gpt-4.1\": { input: 2.0, cachedInput: 0.50, output: 8.0 },\n \"gpt-4.1-mini\": { input: 0.40, cachedInput: 0.10, output: 1.60 },\n \"gpt-4.1-nano\": { input: 0.10, cachedInput: 0.025, output: 0.40 },\n\n // GPT-4o series\n \"gpt-4o\": { input: 2.50, cachedInput: 1.25, output: 10.0 },\n \"gpt-4o-mini\": { input: 0.15, cachedInput: 0.075, output: 0.60 },\n\n // Reasoning models\n \"o3\": { input: 2.0, cachedInput: 0.50, output: 8.0 },\n \"o4-mini\": { input: 1.10, cachedInput: 0.275, output: 4.40 },\n};\n\n// Tool pricing in USD per call\nexport const TOOL_PRICES: Record = {\n \"web_search\": 0.01, // $10 per 1K calls\n \"file_search\": 0.0025, // $2.50 per 1K calls\n \"code_interpreter\": 0.03, // $0.03 per session\n};\n\n// ============================================================================\n// Core Implementation\n// ============================================================================\n\n/**\n * Extract model name from RunResult.\n * SDK stores it at state._currentAgent.model\n */\nfunction getModel(runResult: any, fallback?: string): string {\n return runResult?.state?._currentAgent?.model ?? fallback ?? \"unknown\";\n}\n\n/**\n * Extract and aggregate usage from all model responses.\n * SDK stores responses at state._modelResponses with usage in camelCase.\n */\nfunction getUsage(runResult: any): { input: number; output: number; cached: number } {\n const responses: any[] = runResult?.state?._modelResponses ?? [];\n \n let input = 0;\n let output = 0;\n let cached = 0;\n\n for (const response of responses) {\n const usage = response?.usage;\n if (!usage) continue;\n\n input += usage.inputTokens ?? 0;\n output += usage.outputTokens ?? 0;\n\n // inputTokensDetails is an array with objects like { cachedTokens: N }\n const details = usage.inputTokensDetails;\n if (Array.isArray(details)) {\n for (const d of details) {\n cached += d?.cachedTokens ?? 0;\n }\n }\n }\n\n return { input, output, cached };\n}\n\n/**\n * Extract tool call counts from generated items.\n * SDK stores them at state._generatedItems with type \"tool_call_item\".\n */\nfunction getToolCalls(runResult: any): Record {\n const counts: Record = {};\n const items: any[] = runResult?.state?._generatedItems ?? [];\n\n for (const item of items) {\n if (item?.type !== \"tool_call_item\") continue;\n\n const rawName = item?.rawItem?.name ?? item?.rawItem?.type ?? \"\";\n const normalized = normalizeToolName(rawName);\n if (normalized) {\n counts[normalized] = (counts[normalized] ?? 0) + 1;\n }\n }\n\n return counts;\n}\n\n/**\n * Normalize tool names to match pricing keys.\n * SDK uses names like \"web_search_call\" -> \"web_search\"\n */\nfunction normalizeToolName(name: string): string {\n const lower = name.toLowerCase();\n \n if (lower.includes(\"web_search\")) return \"web_search\";\n if (lower.includes(\"file_search\")) return \"file_search\";\n if (lower.includes(\"code_interpreter\")) return \"code_interpreter\";\n \n // Return cleaned name for custom tools\n return lower.replace(/_call$/, \"\").trim();\n}\n\n/**\n * Calculate USD cost from token count and price per 1M tokens.\n */\nfunction tokenCost(tokens: number, pricePer1M: number): number {\n return (tokens / 1_000_000) * pricePer1M;\n}\n\n// ============================================================================\n// Main Export\n// ============================================================================\n\n/**\n * Estimate the cost of an @openai/agents SDK RunResult.\n */\nexport function estimateCost(runResult: any, options: EstimateOptions = {}): CostBreakdown {\n const modelPrices = { ...MODEL_PRICES, ...options.modelPrices };\n const toolPrices = { ...TOOL_PRICES, ...options.toolPrices };\n\n // Extract data from run result\n const model = getModel(runResult, options.defaultModel);\n const usage = getUsage(runResult);\n const toolCalls = getToolCalls(runResult);\n\n // Get pricing for model\n const pricing = modelPrices[model];\n if (!pricing) {\n throw new Error(\n `No pricing for model \"${model}\". Pass it via options.modelPrices or use options.defaultModel.`\n );\n }\n\n // Calculate token costs\n const nonCachedInput = Math.max(0, usage.input - usage.cached);\n const tokenCostUSD =\n tokenCost(nonCachedInput, pricing.input) +\n tokenCost(usage.cached, pricing.cachedInput) +\n tokenCost(usage.output, pricing.output);\n\n // Calculate tool costs\n let toolCostUSD = 0;\n for (const [tool, count] of Object.entries(toolCalls)) {\n const price = toolPrices[tool];\n if (typeof price === \"number\") {\n toolCostUSD += price * count;\n }\n }\n\n return {\n model,\n tokens: {\n input: usage.input,\n cachedInput: usage.cached,\n nonCachedInput,\n output: usage.output,\n },\n tokenCostUSD,\n toolCalls,\n toolCostUSD,\n totalCostUSD: tokenCostUSD + toolCostUSD,\n };\n}\n\n// ============================================================================\n// Tests (run this file directly to execute)\n// ============================================================================\n\nimport { expect } from \"jsr:@std/expect\";\n\n// Mock data matching actual SDK output from logs\nconst mockSimpleRun = {\n state: {\n _currentAgent: { model: \"gpt-5\" },\n _modelResponses: [{\n usage: {\n inputTokens: 15,\n outputTokens: 7,\n totalTokens: 22,\n inputTokensDetails: [{ cachedTokens: 0 }],\n outputTokensDetails: [{ reasoningTokens: 0 }],\n }\n }],\n _generatedItems: [\n { type: \"reasoning_item\", rawItem: { type: \"reasoning\" } },\n { type: \"message_output_item\", rawItem: { type: \"message\" } },\n ]\n }\n};\n\nconst mockWebSearchRun = {\n state: {\n _currentAgent: { model: \"gpt-5\" },\n _modelResponses: [{\n usage: {\n inputTokens: 69153,\n outputTokens: 3858,\n totalTokens: 73011,\n inputTokensDetails: [{ cachedTokens: 1000 }],\n outputTokensDetails: [{ reasoningTokens: 0 }],\n }\n }],\n _generatedItems: [\n { type: \"reasoning_item\", rawItem: { type: \"reasoning\" } },\n { type: \"tool_call_item\", rawItem: { type: \"hosted_tool_call\", name: \"web_search_call\", status: \"completed\" } },\n { type: \"reasoning_item\", rawItem: { type: \"reasoning\" } },\n { type: \"message_output_item\", rawItem: { type: \"message\" } },\n ]\n }\n};\n\nconst mockMultiResponseRun = {\n state: {\n _currentAgent: { model: \"gpt-4.1-mini\" },\n _modelResponses: [\n { usage: { inputTokens: 100, outputTokens: 50, inputTokensDetails: [{ cachedTokens: 20 }] } },\n { usage: { inputTokens: 200, outputTokens: 100, inputTokensDetails: [{ cachedTokens: 30 }] } },\n ],\n _generatedItems: []\n }\n};\n\nfunction runTests() {\n console.log(\"Running cost2.ts tests...\\n\");\n let passed = 0;\n let failed = 0;\n\n function test(name: string, fn: () => void) {\n try {\n fn();\n console.log(`✓ ${name}`);\n passed++;\n } catch (e) {\n console.log(`✗ ${name}`);\n console.log(` ${e}`);\n failed++;\n }\n }\n\n // Test 1: Simple run without tools\n test(\"simple run - extracts model correctly\", () => {\n const result = estimateCost(mockSimpleRun);\n expect(result.model).toBe(\"gpt-5\");\n });\n\n test(\"simple run - extracts tokens correctly\", () => {\n const result = estimateCost(mockSimpleRun);\n expect(result.tokens.input).toBe(15);\n expect(result.tokens.output).toBe(7);\n expect(result.tokens.cachedInput).toBe(0);\n expect(result.tokens.nonCachedInput).toBe(15);\n });\n\n test(\"simple run - calculates token cost correctly\", () => {\n const result = estimateCost(mockSimpleRun);\n // 15 input tokens at $2.50/1M + 7 output tokens at $20/1M\n const expected = (15 / 1_000_000) * 2.50 + (7 / 1_000_000) * 20.0;\n expect(result.tokenCostUSD).toBeCloseTo(expected, 10);\n });\n\n test(\"simple run - no tool calls\", () => {\n const result = estimateCost(mockSimpleRun);\n expect(Object.keys(result.toolCalls).length).toBe(0);\n expect(result.toolCostUSD).toBe(0);\n });\n\n // Test 2: Run with web search\n test(\"web search run - detects tool call\", () => {\n const result = estimateCost(mockWebSearchRun);\n expect(result.toolCalls[\"web_search\"]).toBe(1);\n });\n\n test(\"web search run - calculates tool cost\", () => {\n const result = estimateCost(mockWebSearchRun);\n expect(result.toolCostUSD).toBe(0.01); // $10/1K = $0.01 per call\n });\n\n test(\"web search run - handles cached tokens\", () => {\n const result = estimateCost(mockWebSearchRun);\n expect(result.tokens.cachedInput).toBe(1000);\n expect(result.tokens.nonCachedInput).toBe(69153 - 1000);\n });\n\n test(\"web search run - total cost includes tools\", () => {\n const result = estimateCost(mockWebSearchRun);\n expect(result.totalCostUSD).toBe(result.tokenCostUSD + result.toolCostUSD);\n });\n\n // Test 3: Multi-response aggregation\n test(\"multi-response - aggregates usage\", () => {\n const result = estimateCost(mockMultiResponseRun);\n expect(result.tokens.input).toBe(300); // 100 + 200\n expect(result.tokens.output).toBe(150); // 50 + 100\n expect(result.tokens.cachedInput).toBe(50); // 20 + 30\n });\n\n // Test 4: Custom model pricing\n test(\"custom model pricing - uses override\", () => {\n const customRun = {\n state: {\n _currentAgent: { model: \"custom-model\" },\n _modelResponses: [{ usage: { inputTokens: 1000, outputTokens: 500, inputTokensDetails: [] } }],\n _generatedItems: []\n }\n };\n \n const result = estimateCost(customRun, {\n modelPrices: { \"custom-model\": { input: 1.0, cachedInput: 0.1, output: 2.0 } }\n });\n \n expect(result.model).toBe(\"custom-model\");\n const expected = (1000 / 1_000_000) * 1.0 + (500 / 1_000_000) * 2.0;\n expect(result.tokenCostUSD).toBeCloseTo(expected, 10);\n });\n\n // Test 5: Default model fallback\n test(\"default model - uses fallback for unknown model\", () => {\n const unknownRun = {\n state: {\n _currentAgent: { model: \"unknown-model-xyz\" },\n _modelResponses: [{ usage: { inputTokens: 100, outputTokens: 50, inputTokensDetails: [] } }],\n _generatedItems: []\n }\n };\n \n const result = estimateCost(unknownRun, { defaultModel: \"gpt-4.1-mini\" });\n // Should use gpt-4.1-mini pricing even though model name is \"unknown-model-xyz\"\n // Wait - this test is wrong. defaultModel is only used if model is missing, not unknown.\n // Let's test the error case instead\n });\n\n // Test 6: Unknown model throws error\n test(\"unknown model - throws helpful error\", () => {\n const unknownRun = {\n state: {\n _currentAgent: { model: \"totally-unknown\" },\n _modelResponses: [{ usage: { inputTokens: 100, outputTokens: 50, inputTokensDetails: [] } }],\n _generatedItems: []\n }\n };\n \n let threw = false;\n try {\n estimateCost(unknownRun);\n } catch (e: any) {\n threw = true;\n expect(e.message).toContain(\"totally-unknown\");\n }\n expect(threw).toBe(true);\n });\n\n // Test 7: Empty/missing data handling\n test(\"empty run - handles gracefully\", () => {\n const emptyRun = {\n state: {\n _currentAgent: { model: \"gpt-5\" },\n _modelResponses: [],\n _generatedItems: []\n }\n };\n \n const result = estimateCost(emptyRun);\n expect(result.tokens.input).toBe(0);\n expect(result.tokens.output).toBe(0);\n expect(result.tokenCostUSD).toBe(0);\n expect(result.totalCostUSD).toBe(0);\n });\n\n // Test 8: Multiple tool calls\n test(\"multiple tool calls - counts correctly\", () => {\n const multiToolRun = {\n state: {\n _currentAgent: { model: \"gpt-5\" },\n _modelResponses: [{ usage: { inputTokens: 100, outputTokens: 50, inputTokensDetails: [] } }],\n _generatedItems: [\n { type: \"tool_call_item\", rawItem: { name: \"web_search_call\" } },\n { type: \"tool_call_item\", rawItem: { name: \"web_search_call\" } },\n { type: \"tool_call_item\", rawItem: { name: \"file_search_call\" } },\n ]\n }\n };\n \n const result = estimateCost(multiToolRun);\n expect(result.toolCalls[\"web_search\"]).toBe(2);\n expect(result.toolCalls[\"file_search\"]).toBe(1);\n expect(result.toolCostUSD).toBeCloseTo(0.01 * 2 + 0.0025 * 1, 10);\n });\n\n // Summary\n console.log(`\\n${passed} passed, ${failed} failed`);\n \n if (failed > 0) {\n throw new Error(`${failed} test(s) failed`);\n }\n}\n\n// Run tests when file is executed directly\nrunTests();\n",[160],"b4bb1ff2-efdc-11f0-9c33-42dde27851f2","cost.ts","8b1cee96-efdc-11f0-a0b9-42dde27851f2",12,["D",1768239165241],2,"/**\n * Cost estimation for @openai/agents SDK runs.\n * Simplified implementation targeting the SDK's actual structure.\n */\n\n// ============================================================================\n// Types\n// ============================================================================\n\ntype ModelPricing = {\n input: number; // $ per 1M input tokens\n cachedInput: number; // $ per 1M cached input tokens\n output: number; // $ per 1M output tokens\n};\n\ntype CostBreakdown = {\n model: string;\n tokens: {\n input: number;\n cachedInput: number;\n nonCachedInput: number;\n output: number;\n };\n tokenCostUSD: number;\n toolCalls: Record;\n toolCostUSD: number;\n totalCostUSD: number;\n};\n\ntype EstimateOptions = {\n modelPrices?: Record;\n toolPrices?: Record;\n defaultModel?: string;\n};\n\n// ============================================================================\n// Default Pricing (January 2026)\n// ============================================================================\n\nexport const MODEL_PRICES: Record = {\n // GPT-5 series\n \"gpt-5\": { input: 2.50, cachedInput: 0.25, output: 20.0 },\n \"gpt-5-mini\": { input: 0.25, cachedInput: 0.025, output: 2.0 },\n\n // GPT-4.1 series\n \"gpt-4.1\": { input: 2.0, cachedInput: 0.50, output: 8.0 },\n \"gpt-4.1-mini\": { input: 0.40, cachedInput: 0.10, output: 1.60 },\n \"gpt-4.1-nano\": { input: 0.10, cachedInput: 0.025, output: 0.40 },\n\n // GPT-4o series\n \"gpt-4o\": { input: 2.50, cachedInput: 1.25, output: 10.0 },\n \"gpt-4o-mini\": { input: 0.15, cachedInput: 0.075, output: 0.60 },\n\n // Reasoning models\n \"o3\": { input: 2.0, cachedInput: 0.50, output: 8.0 },\n \"o4-mini\": { input: 1.10, cachedInput: 0.275, output: 4.40 },\n\n // Computer use\n \"computer-use-preview\": { input: 3.0, cachedInput: 0.0, output: 12.0 },\n};\n\n// Tool pricing in USD per call\nexport const TOOL_PRICES: Record = {\n \"web_search\": 0.01, // $10 per 1K calls\n \"file_search\": 0.0025, // $2.50 per 1K calls\n \"code_interpreter_session\": 0.03, // $0.03 per session\n};\n\n// ============================================================================\n// Core Implementation\n// ============================================================================\n\n/**\n * Extracts the model name from a RunResult, normalizing versioned names.\n */\nfunction getModel(runResult: any): string {\n const model = runResult?.state?._currentAgent?.model;\n if (typeof model !== \"string\") return \"unknown\";\n \n // Normalize \"gpt-5-2025-08-07\" → \"gpt-5\"\n return model.replace(/-\\d{4}-\\d{2}-\\d{2}$/, \"\");\n}\n\n/**\n * Extracts aggregated token usage from all model responses.\n */\nfunction getUsage(runResult: any): { input: number; cached: number; output: number } {\n const responses: any[] = runResult?.state?._modelResponses ?? [];\n \n let input = 0;\n let cached = 0;\n let output = 0;\n\n for (const response of responses) {\n const usage = response?.usage;\n if (!usage) continue;\n\n input += usage.inputTokens ?? 0;\n output += usage.outputTokens ?? 0;\n\n // inputTokensDetails is an array with objects like { cachedTokens: N }\n const details = usage.inputTokensDetails;\n if (Array.isArray(details)) {\n for (const d of details) {\n cached += d?.cachedTokens ?? 0;\n }\n }\n }\n\n return { input, cached, output };\n}\n\n/**\n * Counts tool calls from generated items.\n */\nfunction getToolCalls(runResult: any): Record {\n const counts: Record = {};\n const items: any[] = runResult?.state?._generatedItems ?? [];\n\n for (const item of items) {\n if (item?.type !== \"tool_call_item\") continue;\n\n const rawName = item?.rawItem?.name ?? item?.rawItem?.type ?? \"\";\n const normalized = normalizeToolName(rawName);\n if (normalized) {\n counts[normalized] = (counts[normalized] ?? 0) + 1;\n }\n }\n\n return counts;\n}\n\n/**\n * Normalizes tool names from the SDK format.\n */\nfunction normalizeToolName(name: string): string {\n const s = name.toLowerCase();\n \n if (s.includes(\"web_search\")) return \"web_search\";\n if (s.includes(\"file_search\")) return \"file_search\";\n if (s.includes(\"code_interpreter\")) return \"code_interpreter_session\";\n \n // Strip \"_call\" suffix for unknown tools\n return name.replace(/_call$/i, \"\").trim();\n}\n\n/**\n * Calculates token cost in USD.\n */\nfunction calculateTokenCost(\n tokens: { input: number; cached: number; output: number },\n pricing: ModelPricing\n): number {\n const nonCachedInput = Math.max(0, tokens.input - tokens.cached);\n \n return (\n (nonCachedInput / 1_000_000) * pricing.input +\n (tokens.cached / 1_000_000) * pricing.cachedInput +\n (tokens.output / 1_000_000) * pricing.output\n );\n}\n\n/**\n * Calculates tool cost in USD.\n */\nfunction calculateToolCost(\n toolCalls: Record,\n toolPrices: Record\n): number {\n let cost = 0;\n for (const [tool, count] of Object.entries(toolCalls)) {\n const price = toolPrices[tool];\n if (typeof price === \"number\") {\n cost += price * count;\n }\n }\n return cost;\n}\n\n// ============================================================================\n// Public API\n// ============================================================================\n\n/**\n * Estimates the cost of an @openai/agents SDK RunResult.\n */\nexport function estimateCost(runResult: any, options: EstimateOptions = {}): CostBreakdown {\n const modelPrices = { ...MODEL_PRICES, ...options.modelPrices };\n const toolPrices = { ...TOOL_PRICES, ...options.toolPrices };\n\n const model = getModel(runResult) || options.defaultModel || \"unknown\";\n const usage = getUsage(runResult);\n const toolCalls = getToolCalls(runResult);\n\n const pricing = modelPrices[model];\n if (!pricing) {\n throw new Error(\n `No pricing for model \"${model}\". Add it via options.modelPrices.`\n );\n }\n\n const tokenCostUSD = calculateTokenCost(usage, pricing);\n const toolCostUSD = calculateToolCost(toolCalls, toolPrices);\n\n return {\n model,\n tokens: {\n input: usage.input,\n cachedInput: usage.cached,\n nonCachedInput: Math.max(0, usage.input - usage.cached),\n output: usage.output,\n },\n tokenCostUSD,\n toolCalls,\n toolCostUSD,\n totalCostUSD: tokenCostUSD + toolCostUSD,\n };\n}",[167],"c3158c04-efdc-11f0-a5ad-42dde27851f2",["D",1768239189321],3,[28],"62ae120a-efd7-11f0-8671-42dde27851f2","README.md","019bb321-90d8-7149-add7-4c4cd87898a9",1,["D",1768236880010],"# OpenAI Agents Template\n\nA Val Town template for using the OpenAI Agents SDK.\n\n## Get Started\n\n1. **Remix this val**\n1. **Set the `OPENAI_API_KEY` environment variable.** Get your OpenAI API key\n here: \n1. **Customize your agent:** Edit the agent configuration in `main.ts`. You can\n provide custom `instructions`, choose a different model, or add custom tools.\n1. **Use your agent:** Click `Run` on [`main.ts`](./main.ts)\n\n## Learn More\n\n- [OpenAI Agents SDK Documentation](https://openai.github.io/openai-agents-js/)\n- [OpenAI Agents Quickstart](https://openai.github.io/openai-agents-js/guides/quickstart/)\n- [Val Town Documentation](https://docs.val.town/)",[179],"pulls","branches","forks","envVars",{"_192":181,"_194":63},"connections",{"_192":63,"_193":63},"count","unfilledCount","emptyCount",{"_43":178,"_32":179,"_30":180,"_150":-5,"_41":42,"_40":181,"_152":153,"_38":213,"_13":63,"_33":20,"_35":183,"_24":25,"_157":-5,"_201":214,"_27":215},{"_43":174,"_32":28,"_30":31,"_150":-5,"_41":42,"_40":14,"_152":153,"_38":210,"_13":176,"_33":34,"_35":37,"_24":25,"_157":-5,"_201":211,"_27":212},{"_43":166,"_32":167,"_30":168,"_150":-5,"_41":42,"_40":169,"_152":25,"_38":207,"_13":171,"_33":34,"_35":172,"_24":25,"_157":-5,"_201":208,"_27":209},{"_43":159,"_32":160,"_30":161,"_150":-5,"_41":42,"_40":162,"_152":25,"_38":204,"_13":63,"_33":34,"_35":164,"_24":25,"_157":-5,"_201":205,"_27":206},{"_43":148,"_32":149,"_30":58,"_150":-5,"_41":42,"_40":151,"_152":153,"_38":200,"_13":155,"_33":34,"_35":156,"_24":25,"_157":-5,"_201":202,"_27":203},["D",1768236968140],"children",[],[149],["D",1768239168920],[],[160],["D",1768239165241],[],[167],["D",1768239189321],[],[28],["D",1768236880010],[],[179],["D",1768236880010],{"_101":100,"_108":104},{"_101":100,"_102":-5,"_103":104,"_18":56,"_105":106,"_107":25},{"_13":86,"_32":87,"_43":42,"_88":44,"_89":90,"_91":92,"_93":50,"_94":95,"_96":97,"_98":45,"_99":100},"ctx",{"_79":-5,"_130":253,"_254":255,"_256":257,"_258":25,"_259":25,"_260":25,"_261":153,"_262":-5,"_263":-7,"_264":-7,"_265":63,"_266":25,"_267":25,"_268":-5,"_269":25,"_270":25,"_271":25,"_272":-5},"ENV",{"_224":225,"_226":227,"_228":87,"_229":230,"_231":232,"_233":234,"_235":236,"_237":238,"_239":240,"_241":242,"_243":244,"_245":246,"_247":248,"_249":250,"_251":252},"REACT_APP_SENTRY_DSN","https://586020d86f05489db63a554690acf1e9@o1398359.ingest.sentry.io/6724499","RENDER_GIT_COMMIT","5015b8df46402d8bba4f485ecc37aec202b0e06b","RENDER_GIT_BRANCH","POSTHOG_API_TOKEN","phc_nmoyuAkStEZuUp69rYlg1DZgrIt19Gx0Y4Hwf8tGgJP","POSTHOG_API_HOST","https://wart.val.town","WEB_API_TEMPLATE","https://{val}.web.val.run","APP_URL","https://www.val.town","MODULE_URL","https://esm.town","VANITY_WEB_API_TEMPLATE","https://{label}.val.run/","DENOLS_URL","https://lsp.val.town","NODE_ENV","production","CLOUDFLARE_TURNSTILE_SITE_KEY","0x4AAAAAACtaIZJH7s9rBMZ6","GOOGLE_CLIENT_ID","875989696899-jtsqkhlbdv5jrlsk8631p7ipjg3c6o62.apps.googleusercontent.com","STRIPE_PUBLISHABLE_KEY","pk_live_51N2zvWEpZTTLipIkVEwXrUG9AT0yeOOEYxszarZT1jpsZV58Ly6wlg3SltZpHN26CUalOBmUSY7adQRQiSkCZz2500qSXNkf8Y","API_URL","https://api.val.town",[],"preferencesKeybindings","default","preferencesTheme","light","preferencesFixers","preferencesCopilot","preferencesSelectionAi","preferencesCloseBrackets","preferencesChatPanelState","preferencesDprint","preferencesDenofmt","notificationCount","flagDenols","isBusinessOrgMember","businessOrgHandle","isBusinessOrgAdmin","isHigherTownieTier","navCollapsed","currentOrgId"]

Diff · v17