nodebench-mcp 3.0.0 → 3.0.2
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/NODEBENCH_AGENTS.md +74 -67
- package/README.md +36 -34
- package/dist/dashboard/operatingDashboardHtml.js +2 -1
- package/dist/dashboard/operatingDashboardHtml.js.map +1 -1
- package/dist/dashboard/operatingServer.js +3 -2
- package/dist/dashboard/operatingServer.js.map +1 -1
- package/dist/db.js +51 -3
- package/dist/db.js.map +1 -1
- package/dist/index.js +19 -18
- package/dist/index.js.map +1 -1
- package/dist/packageInfo.d.ts +3 -0
- package/dist/packageInfo.js +32 -0
- package/dist/packageInfo.js.map +1 -0
- package/dist/sandboxApi.js +2 -1
- package/dist/sandboxApi.js.map +1 -1
- package/dist/tools/boilerplateTools.js +10 -9
- package/dist/tools/boilerplateTools.js.map +1 -1
- package/dist/tools/documentationTools.js +2 -1
- package/dist/tools/documentationTools.js.map +1 -1
- package/dist/tools/progressiveDiscoveryTools.js +2 -1
- package/dist/tools/progressiveDiscoveryTools.js.map +1 -1
- package/dist/tools/toolRegistry.js +11 -0
- package/dist/tools/toolRegistry.js.map +1 -1
- package/dist/toolsetRegistry.js +74 -1
- package/dist/toolsetRegistry.js.map +1 -1
- package/package.json +7 -6
- package/scripts/install.sh +14 -14
- package/dist/__tests__/analytics.test.d.ts +0 -11
- package/dist/__tests__/analytics.test.js +0 -546
- package/dist/__tests__/analytics.test.js.map +0 -1
- package/dist/__tests__/architectComplex.test.d.ts +0 -1
- package/dist/__tests__/architectComplex.test.js +0 -373
- package/dist/__tests__/architectComplex.test.js.map +0 -1
- package/dist/__tests__/architectSmoke.test.d.ts +0 -1
- package/dist/__tests__/architectSmoke.test.js +0 -92
- package/dist/__tests__/architectSmoke.test.js.map +0 -1
- package/dist/__tests__/audit-registry.d.ts +0 -1
- package/dist/__tests__/audit-registry.js +0 -60
- package/dist/__tests__/audit-registry.js.map +0 -1
- package/dist/__tests__/batchAutopilot.test.d.ts +0 -8
- package/dist/__tests__/batchAutopilot.test.js +0 -218
- package/dist/__tests__/batchAutopilot.test.js.map +0 -1
- package/dist/__tests__/cliSubcommands.test.d.ts +0 -1
- package/dist/__tests__/cliSubcommands.test.js +0 -138
- package/dist/__tests__/cliSubcommands.test.js.map +0 -1
- package/dist/__tests__/comparativeBench.test.d.ts +0 -1
- package/dist/__tests__/comparativeBench.test.js +0 -722
- package/dist/__tests__/comparativeBench.test.js.map +0 -1
- package/dist/__tests__/critterCalibrationEval.d.ts +0 -8
- package/dist/__tests__/critterCalibrationEval.js +0 -370
- package/dist/__tests__/critterCalibrationEval.js.map +0 -1
- package/dist/__tests__/dynamicLoading.test.d.ts +0 -1
- package/dist/__tests__/dynamicLoading.test.js +0 -280
- package/dist/__tests__/dynamicLoading.test.js.map +0 -1
- package/dist/__tests__/embeddingProvider.test.d.ts +0 -1
- package/dist/__tests__/embeddingProvider.test.js +0 -86
- package/dist/__tests__/embeddingProvider.test.js.map +0 -1
- package/dist/__tests__/evalDatasetBench.test.d.ts +0 -1
- package/dist/__tests__/evalDatasetBench.test.js +0 -738
- package/dist/__tests__/evalDatasetBench.test.js.map +0 -1
- package/dist/__tests__/evalHarness.test.d.ts +0 -1
- package/dist/__tests__/evalHarness.test.js +0 -1107
- package/dist/__tests__/evalHarness.test.js.map +0 -1
- package/dist/__tests__/fixtures/bfcl_v3_long_context.sample.json +0 -264
- package/dist/__tests__/fixtures/generateBfclLongContextFixture.d.ts +0 -10
- package/dist/__tests__/fixtures/generateBfclLongContextFixture.js +0 -135
- package/dist/__tests__/fixtures/generateBfclLongContextFixture.js.map +0 -1
- package/dist/__tests__/fixtures/generateSwebenchVerifiedFixture.d.ts +0 -14
- package/dist/__tests__/fixtures/generateSwebenchVerifiedFixture.js +0 -189
- package/dist/__tests__/fixtures/generateSwebenchVerifiedFixture.js.map +0 -1
- package/dist/__tests__/fixtures/generateToolbenchInstructionFixture.d.ts +0 -16
- package/dist/__tests__/fixtures/generateToolbenchInstructionFixture.js +0 -154
- package/dist/__tests__/fixtures/generateToolbenchInstructionFixture.js.map +0 -1
- package/dist/__tests__/fixtures/swebench_verified.sample.json +0 -162
- package/dist/__tests__/fixtures/toolbench_instruction.sample.json +0 -109
- package/dist/__tests__/forecastingDogfood.test.d.ts +0 -9
- package/dist/__tests__/forecastingDogfood.test.js +0 -284
- package/dist/__tests__/forecastingDogfood.test.js.map +0 -1
- package/dist/__tests__/forecastingScoring.test.d.ts +0 -9
- package/dist/__tests__/forecastingScoring.test.js +0 -202
- package/dist/__tests__/forecastingScoring.test.js.map +0 -1
- package/dist/__tests__/gaiaCapabilityAudioEval.test.d.ts +0 -15
- package/dist/__tests__/gaiaCapabilityAudioEval.test.js +0 -265
- package/dist/__tests__/gaiaCapabilityAudioEval.test.js.map +0 -1
- package/dist/__tests__/gaiaCapabilityEval.test.d.ts +0 -14
- package/dist/__tests__/gaiaCapabilityEval.test.js +0 -1259
- package/dist/__tests__/gaiaCapabilityEval.test.js.map +0 -1
- package/dist/__tests__/gaiaCapabilityFilesEval.test.d.ts +0 -15
- package/dist/__tests__/gaiaCapabilityFilesEval.test.js +0 -914
- package/dist/__tests__/gaiaCapabilityFilesEval.test.js.map +0 -1
- package/dist/__tests__/gaiaCapabilityMediaEval.test.d.ts +0 -15
- package/dist/__tests__/gaiaCapabilityMediaEval.test.js +0 -1101
- package/dist/__tests__/gaiaCapabilityMediaEval.test.js.map +0 -1
- package/dist/__tests__/helpers/answerMatch.d.ts +0 -41
- package/dist/__tests__/helpers/answerMatch.js +0 -267
- package/dist/__tests__/helpers/answerMatch.js.map +0 -1
- package/dist/__tests__/helpers/textLlm.d.ts +0 -25
- package/dist/__tests__/helpers/textLlm.js +0 -214
- package/dist/__tests__/helpers/textLlm.js.map +0 -1
- package/dist/__tests__/localDashboard.test.d.ts +0 -1
- package/dist/__tests__/localDashboard.test.js +0 -226
- package/dist/__tests__/localDashboard.test.js.map +0 -1
- package/dist/__tests__/multiHopDogfood.test.d.ts +0 -12
- package/dist/__tests__/multiHopDogfood.test.js +0 -303
- package/dist/__tests__/multiHopDogfood.test.js.map +0 -1
- package/dist/__tests__/openDatasetParallelEval.test.d.ts +0 -7
- package/dist/__tests__/openDatasetParallelEval.test.js +0 -209
- package/dist/__tests__/openDatasetParallelEval.test.js.map +0 -1
- package/dist/__tests__/openDatasetParallelEvalGaia.test.d.ts +0 -7
- package/dist/__tests__/openDatasetParallelEvalGaia.test.js +0 -279
- package/dist/__tests__/openDatasetParallelEvalGaia.test.js.map +0 -1
- package/dist/__tests__/openDatasetParallelEvalSwebench.test.d.ts +0 -7
- package/dist/__tests__/openDatasetParallelEvalSwebench.test.js +0 -220
- package/dist/__tests__/openDatasetParallelEvalSwebench.test.js.map +0 -1
- package/dist/__tests__/openDatasetParallelEvalToolbench.test.d.ts +0 -7
- package/dist/__tests__/openDatasetParallelEvalToolbench.test.js +0 -218
- package/dist/__tests__/openDatasetParallelEvalToolbench.test.js.map +0 -1
- package/dist/__tests__/openDatasetPerfComparison.test.d.ts +0 -10
- package/dist/__tests__/openDatasetPerfComparison.test.js +0 -318
- package/dist/__tests__/openDatasetPerfComparison.test.js.map +0 -1
- package/dist/__tests__/openclawDogfood.test.d.ts +0 -23
- package/dist/__tests__/openclawDogfood.test.js +0 -535
- package/dist/__tests__/openclawDogfood.test.js.map +0 -1
- package/dist/__tests__/openclawMessaging.test.d.ts +0 -14
- package/dist/__tests__/openclawMessaging.test.js +0 -232
- package/dist/__tests__/openclawMessaging.test.js.map +0 -1
- package/dist/__tests__/presetRealWorldBench.test.d.ts +0 -1
- package/dist/__tests__/presetRealWorldBench.test.js +0 -859
- package/dist/__tests__/presetRealWorldBench.test.js.map +0 -1
- package/dist/__tests__/tools.test.d.ts +0 -1
- package/dist/__tests__/tools.test.js +0 -3201
- package/dist/__tests__/tools.test.js.map +0 -1
- package/dist/__tests__/toolsetGatingEval.test.d.ts +0 -1
- package/dist/__tests__/toolsetGatingEval.test.js +0 -1099
- package/dist/__tests__/toolsetGatingEval.test.js.map +0 -1
- package/dist/__tests__/traceabilityDogfood.test.d.ts +0 -12
- package/dist/__tests__/traceabilityDogfood.test.js +0 -241
- package/dist/__tests__/traceabilityDogfood.test.js.map +0 -1
- package/dist/__tests__/webmcpTools.test.d.ts +0 -7
- package/dist/__tests__/webmcpTools.test.js +0 -195
- package/dist/__tests__/webmcpTools.test.js.map +0 -1
- package/dist/benchmarks/testProviderBus.d.ts +0 -7
- package/dist/benchmarks/testProviderBus.js +0 -272
- package/dist/benchmarks/testProviderBus.js.map +0 -1
- package/dist/hooks/postCompaction.d.ts +0 -14
- package/dist/hooks/postCompaction.js +0 -51
- package/dist/hooks/postCompaction.js.map +0 -1
- package/dist/security/__tests__/security.test.d.ts +0 -8
- package/dist/security/__tests__/security.test.js +0 -295
- package/dist/security/__tests__/security.test.js.map +0 -1
- package/dist/sync/hyperloopEval.test.d.ts +0 -4
- package/dist/sync/hyperloopEval.test.js +0 -60
- package/dist/sync/hyperloopEval.test.js.map +0 -1
- package/dist/sync/store.test.d.ts +0 -4
- package/dist/sync/store.test.js +0 -43
- package/dist/sync/store.test.js.map +0 -1
- package/dist/tools/documentTools.d.ts +0 -5
- package/dist/tools/documentTools.js +0 -524
- package/dist/tools/documentTools.js.map +0 -1
- package/dist/tools/financialTools.d.ts +0 -10
- package/dist/tools/financialTools.js +0 -403
- package/dist/tools/financialTools.js.map +0 -1
- package/dist/tools/memoryTools.d.ts +0 -5
- package/dist/tools/memoryTools.js +0 -137
- package/dist/tools/memoryTools.js.map +0 -1
- package/dist/tools/planningTools.d.ts +0 -5
- package/dist/tools/planningTools.js +0 -147
- package/dist/tools/planningTools.js.map +0 -1
- package/dist/tools/searchTools.d.ts +0 -5
- package/dist/tools/searchTools.js +0 -145
- package/dist/tools/searchTools.js.map +0 -1
|
@@ -1,220 +0,0 @@
|
|
|
1
|
-
/**
|
|
2
|
-
* Open-source dataset benchmark for long-running tasks (SWE-bench lane).
|
|
3
|
-
*
|
|
4
|
-
* This test uses SWE-bench Verified issue tasks and runs task workflows
|
|
5
|
-
* through NodeBench MCP tools in parallel "subagent" workers.
|
|
6
|
-
*/
|
|
7
|
-
import { describe, expect, it } from "vitest";
|
|
8
|
-
import datasetFixture from "./fixtures/swebench_verified.sample.json";
|
|
9
|
-
import { verificationTools } from "../tools/verificationTools.js";
|
|
10
|
-
import { reconTools } from "../tools/reconTools.js";
|
|
11
|
-
import { evalTools } from "../tools/evalTools.js";
|
|
12
|
-
import { qualityGateTools } from "../tools/qualityGateTools.js";
|
|
13
|
-
import { flywheelTools } from "../tools/flywheelTools.js";
|
|
14
|
-
import { learningTools } from "../tools/learningTools.js";
|
|
15
|
-
import { documentationTools } from "../tools/documentationTools.js";
|
|
16
|
-
import { agentBootstrapTools } from "../tools/agentBootstrapTools.js";
|
|
17
|
-
import { createMetaTools } from "../tools/metaTools.js";
|
|
18
|
-
const fixture = datasetFixture;
|
|
19
|
-
const domainTools = [
|
|
20
|
-
...verificationTools,
|
|
21
|
-
...evalTools,
|
|
22
|
-
...qualityGateTools,
|
|
23
|
-
...learningTools,
|
|
24
|
-
...flywheelTools,
|
|
25
|
-
...reconTools,
|
|
26
|
-
...documentationTools,
|
|
27
|
-
...agentBootstrapTools,
|
|
28
|
-
];
|
|
29
|
-
const allTools = [...domainTools, ...createMetaTools(domainTools)];
|
|
30
|
-
const openDatasetToolCallLog = [];
|
|
31
|
-
function findTool(name) {
|
|
32
|
-
const tool = allTools.find((candidate) => candidate.name === name);
|
|
33
|
-
if (!tool)
|
|
34
|
-
throw new Error(`Tool not found: ${name}`);
|
|
35
|
-
return tool;
|
|
36
|
-
}
|
|
37
|
-
async function callTool(name, args, taskId, stage) {
|
|
38
|
-
const tool = findTool(name);
|
|
39
|
-
try {
|
|
40
|
-
const result = await tool.handler(args);
|
|
41
|
-
openDatasetToolCallLog.push({ taskId, tool: name, stage, success: true });
|
|
42
|
-
return result;
|
|
43
|
-
}
|
|
44
|
-
catch (error) {
|
|
45
|
-
openDatasetToolCallLog.push({ taskId, tool: name, stage, success: false });
|
|
46
|
-
throw error;
|
|
47
|
-
}
|
|
48
|
-
}
|
|
49
|
-
async function runDatasetTask(task, workerIndex) {
|
|
50
|
-
const started = Date.now();
|
|
51
|
-
const recon = (await callTool("run_recon", {
|
|
52
|
-
target: `SWE-bench Verified task ${task.id}`,
|
|
53
|
-
description: `Open-source long-running engineering benchmark (${task.repo}, ${task.difficulty}).`,
|
|
54
|
-
projectContext: {
|
|
55
|
-
techStack: "TypeScript, MCP, SQLite",
|
|
56
|
-
architecture: "Parallel subagent MCP workflow benchmark",
|
|
57
|
-
},
|
|
58
|
-
}, task.id, "recon_start"));
|
|
59
|
-
await callTool("log_recon_finding", {
|
|
60
|
-
sessionId: recon.sessionId,
|
|
61
|
-
category: "dataset",
|
|
62
|
-
summary: `Ingested SWE-bench task ${task.id} (${task.repo}) with failToPass=${task.failToPassCount}.`,
|
|
63
|
-
sourceUrl: `${fixture.sourceUrl}`,
|
|
64
|
-
relevance: "Real-world software issue benchmark for MCP orchestration quality.",
|
|
65
|
-
actionItems: "Run in parallel worker pool and enforce mandatory flywheel checks.",
|
|
66
|
-
}, task.id, "recon_log");
|
|
67
|
-
let discovered = (await callTool("findTools", {
|
|
68
|
-
query: task.prompt.slice(0, 600),
|
|
69
|
-
category: "verification",
|
|
70
|
-
}, task.id, "find_tools"));
|
|
71
|
-
if (!Array.isArray(discovered?.tools) || discovered.tools.length === 0) {
|
|
72
|
-
discovered = (await callTool("findTools", { query: "software bugfix verification eval workflow", category: "eval" }, task.id, "find_tools_fallback"));
|
|
73
|
-
}
|
|
74
|
-
expect(Array.isArray(discovered.tools)).toBe(true);
|
|
75
|
-
expect(discovered.tools.length).toBeGreaterThan(0);
|
|
76
|
-
const difficultyLower = task.difficulty.toLowerCase();
|
|
77
|
-
const methodologyTopic = difficultyLower.includes("1-4 hours") || task.complexityScore >= 140
|
|
78
|
-
? "mandatory_flywheel"
|
|
79
|
-
: "verification";
|
|
80
|
-
const methodology = (await callTool("getMethodology", { topic: methodologyTopic }, task.id, "get_methodology"));
|
|
81
|
-
expect(methodology.title).toBeTruthy();
|
|
82
|
-
expect(Array.isArray(methodology.steps)).toBe(true);
|
|
83
|
-
expect(methodology.steps.length).toBeGreaterThan(0);
|
|
84
|
-
const evalRun = (await callTool("start_eval_run", {
|
|
85
|
-
name: `open-dataset-swebench-${task.id}-${Date.now()}`,
|
|
86
|
-
description: `SWE-bench scenario (${task.repo}, worker ${workerIndex})`,
|
|
87
|
-
cases: [
|
|
88
|
-
{
|
|
89
|
-
input: task.prompt,
|
|
90
|
-
intent: `Coordinate long-running SWE-bench workflow for ${task.id}`,
|
|
91
|
-
expected: "Discover strategy, run eval bookkeeping, and complete closed-loop plus mandatory flywheel checks.",
|
|
92
|
-
},
|
|
93
|
-
],
|
|
94
|
-
}, task.id, "start_eval_run"));
|
|
95
|
-
await callTool("record_eval_result", {
|
|
96
|
-
caseId: evalRun.caseIds[0],
|
|
97
|
-
verdict: "pass",
|
|
98
|
-
score: 1,
|
|
99
|
-
actual: `Discovered ${discovered.tools.length} tools and completed SWE-bench workflow.`,
|
|
100
|
-
telemetry: {
|
|
101
|
-
dataset: fixture.dataset,
|
|
102
|
-
split: fixture.split,
|
|
103
|
-
taskId: task.id,
|
|
104
|
-
repo: task.repo,
|
|
105
|
-
difficulty: task.difficulty,
|
|
106
|
-
workerIndex,
|
|
107
|
-
statementLength: task.statementLength,
|
|
108
|
-
hintLength: task.hintLength,
|
|
109
|
-
failToPassCount: task.failToPassCount,
|
|
110
|
-
passToPassCount: task.passToPassCount,
|
|
111
|
-
complexityScore: task.complexityScore,
|
|
112
|
-
},
|
|
113
|
-
}, task.id, "record_eval_result");
|
|
114
|
-
const evalSummary = (await callTool("complete_eval_run", { runId: evalRun.runId }, task.id, "complete_eval_run"));
|
|
115
|
-
expect(evalSummary.status).toBe("completed");
|
|
116
|
-
expect(evalSummary.summary.passed).toBe(1);
|
|
117
|
-
const closedLoop = (await callTool("run_closed_loop", {
|
|
118
|
-
steps: [
|
|
119
|
-
{ step: "compile", passed: true, output: `Compile checks for ${task.id}` },
|
|
120
|
-
{ step: "lint", passed: true, output: `Lint checks for ${task.id}` },
|
|
121
|
-
{ step: "test", passed: true, output: `Parallel benchmark checks for ${task.id}` },
|
|
122
|
-
],
|
|
123
|
-
}, task.id, "run_closed_loop"));
|
|
124
|
-
expect(closedLoop.allPassed).toBe(true);
|
|
125
|
-
const flywheel = (await callTool("run_mandatory_flywheel", {
|
|
126
|
-
target: `Open-source SWE-bench task ${task.id}`,
|
|
127
|
-
steps: [
|
|
128
|
-
{ stepName: "static_analysis", passed: true, output: "Types and schemas validated." },
|
|
129
|
-
{ stepName: "happy_path_test", passed: true, output: "Dataset task completed end-to-end." },
|
|
130
|
-
{ stepName: "failure_path_test", passed: true, output: "Fallback discovery query validated." },
|
|
131
|
-
{ stepName: "gap_analysis", passed: true, output: "No blocking gaps for this task." },
|
|
132
|
-
{ stepName: "fix_and_reverify", passed: true, output: "No rework required after checks." },
|
|
133
|
-
{ stepName: "deploy_and_document", passed: true, output: "Benchmark result documented." },
|
|
134
|
-
],
|
|
135
|
-
}, task.id, "run_mandatory_flywheel"));
|
|
136
|
-
expect(flywheel.passed).toBe(true);
|
|
137
|
-
const knowledge = (await callTool("search_all_knowledge", { query: task.id, limit: 10 }, task.id, "search_all_knowledge"));
|
|
138
|
-
expect(typeof knowledge.totalResults).toBe("number");
|
|
139
|
-
expect(knowledge.totalResults).toBeGreaterThan(0);
|
|
140
|
-
return {
|
|
141
|
-
taskId: task.id,
|
|
142
|
-
workerIndex,
|
|
143
|
-
ok: true,
|
|
144
|
-
elapsedMs: Date.now() - started,
|
|
145
|
-
discoveredTools: discovered.tools.length,
|
|
146
|
-
knowledgeHits: knowledge.totalResults,
|
|
147
|
-
};
|
|
148
|
-
}
|
|
149
|
-
async function runWorkerPool(tasks, concurrency) {
|
|
150
|
-
const boundedConcurrency = Math.max(1, Math.min(concurrency, tasks.length));
|
|
151
|
-
const results = new Array(tasks.length);
|
|
152
|
-
let nextIndex = 0;
|
|
153
|
-
const workers = Array.from({ length: boundedConcurrency }, (_, workerIndex) => (async () => {
|
|
154
|
-
while (true) {
|
|
155
|
-
const taskIndex = nextIndex++;
|
|
156
|
-
if (taskIndex >= tasks.length)
|
|
157
|
-
return;
|
|
158
|
-
const task = tasks[taskIndex];
|
|
159
|
-
try {
|
|
160
|
-
results[taskIndex] = await runDatasetTask(task, workerIndex);
|
|
161
|
-
}
|
|
162
|
-
catch (error) {
|
|
163
|
-
results[taskIndex] = {
|
|
164
|
-
taskId: task.id,
|
|
165
|
-
workerIndex,
|
|
166
|
-
ok: false,
|
|
167
|
-
elapsedMs: 0,
|
|
168
|
-
discoveredTools: 0,
|
|
169
|
-
knowledgeHits: 0,
|
|
170
|
-
error: error instanceof Error ? error.message : String(error),
|
|
171
|
-
};
|
|
172
|
-
}
|
|
173
|
-
}
|
|
174
|
-
})());
|
|
175
|
-
await Promise.all(workers);
|
|
176
|
-
return results;
|
|
177
|
-
}
|
|
178
|
-
describe("Scenario: Open-Source Long-Running Dataset (SWE-bench Parallel Subagents)", () => {
|
|
179
|
-
it("should execute SWE-bench tasks with parallel MCP subagent workflows", async () => {
|
|
180
|
-
expect(Array.isArray(fixture.tasks)).toBe(true);
|
|
181
|
-
expect(fixture.tasks.length).toBeGreaterThan(0);
|
|
182
|
-
const requestedTaskLimit = Number.parseInt(process.env.NODEBENCH_SWEBENCH_TASK_LIMIT ?? "8", 10);
|
|
183
|
-
const taskLimit = Math.max(1, Math.min(fixture.tasks.length, Number.isFinite(requestedTaskLimit) ? requestedTaskLimit : 8));
|
|
184
|
-
const requestedConcurrency = Number.parseInt(process.env.NODEBENCH_SWEBENCH_CONCURRENCY ?? "4", 10);
|
|
185
|
-
const concurrency = Math.max(1, Math.min(taskLimit, Number.isFinite(requestedConcurrency) ? requestedConcurrency : 4));
|
|
186
|
-
const tasks = fixture.tasks.slice(0, taskLimit);
|
|
187
|
-
const started = Date.now();
|
|
188
|
-
const results = await runWorkerPool(tasks, concurrency);
|
|
189
|
-
const elapsedMs = Date.now() - started;
|
|
190
|
-
const failed = results.filter((result) => !result.ok);
|
|
191
|
-
const passed = results.filter((result) => result.ok);
|
|
192
|
-
const calledTools = new Set(openDatasetToolCallLog.map((entry) => entry.tool));
|
|
193
|
-
const requiredTools = [
|
|
194
|
-
"run_recon",
|
|
195
|
-
"log_recon_finding",
|
|
196
|
-
"findTools",
|
|
197
|
-
"getMethodology",
|
|
198
|
-
"start_eval_run",
|
|
199
|
-
"record_eval_result",
|
|
200
|
-
"complete_eval_run",
|
|
201
|
-
"run_closed_loop",
|
|
202
|
-
"run_mandatory_flywheel",
|
|
203
|
-
"search_all_knowledge",
|
|
204
|
-
];
|
|
205
|
-
console.log(`[open-dataset-swebench] dataset=${fixture.dataset} split=${fixture.split} tasks=${taskLimit} concurrency=${concurrency} pass=${passed.length}/${results.length} elapsedMs=${elapsedMs} toolCalls=${openDatasetToolCallLog.length}`);
|
|
206
|
-
if (failed.length > 0) {
|
|
207
|
-
console.error("[open-dataset-swebench] failures:", failed.map((result) => ({
|
|
208
|
-
taskId: result.taskId,
|
|
209
|
-
workerIndex: result.workerIndex,
|
|
210
|
-
error: result.error,
|
|
211
|
-
})));
|
|
212
|
-
}
|
|
213
|
-
expect(failed.length).toBe(0);
|
|
214
|
-
expect(passed.length).toBe(taskLimit);
|
|
215
|
-
for (const requiredTool of requiredTools) {
|
|
216
|
-
expect(calledTools.has(requiredTool)).toBe(true);
|
|
217
|
-
}
|
|
218
|
-
});
|
|
219
|
-
});
|
|
220
|
-
//# sourceMappingURL=openDatasetParallelEvalSwebench.test.js.map
|
|
@@ -1 +0,0 @@
|
|
|
1
|
-
{"version":3,"file":"openDatasetParallelEvalSwebench.test.js","sourceRoot":"","sources":["../../src/__tests__/openDatasetParallelEvalSwebench.test.ts"],"names":[],"mappings":"AAAA;;;;;GAKG;AAEH,OAAO,EAAE,QAAQ,EAAE,MAAM,EAAE,EAAE,EAAE,MAAM,QAAQ,CAAC;AAC9C,OAAO,cAAc,MAAM,0CAA0C,CAAC;AACtE,OAAO,EAAE,iBAAiB,EAAE,MAAM,+BAA+B,CAAC;AAClE,OAAO,EAAE,UAAU,EAAE,MAAM,wBAAwB,CAAC;AACpD,OAAO,EAAE,SAAS,EAAE,MAAM,uBAAuB,CAAC;AAClD,OAAO,EAAE,gBAAgB,EAAE,MAAM,8BAA8B,CAAC;AAChE,OAAO,EAAE,aAAa,EAAE,MAAM,2BAA2B,CAAC;AAC1D,OAAO,EAAE,aAAa,EAAE,MAAM,2BAA2B,CAAC;AAC1D,OAAO,EAAE,kBAAkB,EAAE,MAAM,gCAAgC,CAAC;AACpE,OAAO,EAAE,mBAAmB,EAAE,MAAM,iCAAiC,CAAC;AACtE,OAAO,EAAE,eAAe,EAAE,MAAM,uBAAuB,CAAC;AA4CxD,MAAM,OAAO,GAAG,cAAgC,CAAC;AAEjD,MAAM,WAAW,GAAc;IAC7B,GAAG,iBAAiB;IACpB,GAAG,SAAS;IACZ,GAAG,gBAAgB;IACnB,GAAG,aAAa;IAChB,GAAG,aAAa;IAChB,GAAG,UAAU;IACb,GAAG,kBAAkB;IACrB,GAAG,mBAAmB;CACvB,CAAC;AACF,MAAM,QAAQ,GAAG,CAAC,GAAG,WAAW,EAAE,GAAG,eAAe,CAAC,WAAW,CAAC,CAAC,CAAC;AAEnE,MAAM,sBAAsB,GAKvB,EAAE,CAAC;AAER,SAAS,QAAQ,CAAC,IAAY;IAC5B,MAAM,IAAI,GAAG,QAAQ,CAAC,IAAI,CAAC,CAAC,SAAS,EAAE,EAAE,CAAC,SAAS,CAAC,IAAI,KAAK,IAAI,CAAC,CAAC;IACnE,IAAI,CAAC,IAAI;QAAE,MAAM,IAAI,KAAK,CAAC,mBAAmB,IAAI,EAAE,CAAC,CAAC;IACtD,OAAO,IAAI,CAAC;AACd,CAAC;AAED,KAAK,UAAU,QAAQ,CACrB,IAAY,EACZ,IAA6B,EAC7B,MAAc,EACd,KAAa;IAEb,MAAM,IAAI,GAAG,QAAQ,CAAC,IAAI,CAAC,CAAC;IAC5B,IAAI,CAAC;QACH,MAAM,MAAM,GAAG,MAAM,IAAI,CAAC,OAAO,CAAC,IAAI,CAAC,CAAC;QACxC,sBAAsB,CAAC,IAAI,CAAC,EAAE,MAAM,EAAE,IAAI,EAAE,IAAI,EAAE,KAAK,EAAE,OAAO,EAAE,IAAI,EAAE,CAAC,CAAC;QAC1E,OAAO,MAAM,CAAC;IAChB,CAAC;IAAC,OAAO,KAAK,EAAE,CAAC;QACf,sBAAsB,CAAC,IAAI,CAAC,EAAE,MAAM,EAAE,IAAI,EAAE,IAAI,EAAE,KAAK,EAAE,OAAO,EAAE,KAAK,EAAE,CAAC,CAAC;QAC3E,MAAM,KAAK,CAAC;IACd,CAAC;AACH,CAAC;AAED,KAAK,UAAU,cAAc,CAAC,IAAiB,EAAE,WAAmB;IAClE,MAAM,OAAO,GAAG,IAAI,CAAC,GAAG,EAAE,CAAC;IAE3B,MAAM,KAAK,GAAG,CAAC,MAAM,QAAQ,CAC3B,WAAW,EACX;QACE,MAAM,EAAE,2BAA2B,IAAI,CAAC,EAAE,EAAE;QAC5C,WAAW,EAAE,mDAAmD,IAAI,CAAC,IAAI,KAAK,IAAI,CAAC,UAAU,IAAI;QACjG,cAAc,EAAE;YACd,SAAS,EAAE,yBAAyB;YACpC,YAAY,EAAE,0CAA0C;SACzD;KACF,EACD,IAAI,CAAC,EAAE,EACP,aAAa,CACd,CAAQ,CAAC;IAEV,MAAM,QAAQ,CACZ,mBAAmB,EACnB;QACE,SAAS,EAAE,KAAK,CAAC,SAAS;QAC1B,QAAQ,EAAE,SAAS;QACnB,OAAO,EAAE,2BAA2B,IAAI,CAAC,EAAE,KAAK,IAAI,CAAC,IAAI,qBAAqB,IAAI,CAAC,eAAe,GAAG;QACrG,SAAS,EAAE,GAAG,OAAO,CAAC,SAAS,EAAE;QACjC,SAAS,EAAE,oEAAoE;QAC/E,WAAW,EAAE,oEAAoE;KAClF,EACD,IAAI,CAAC,EAAE,EACP,WAAW,CACZ,CAAC;IAEF,IAAI,UAAU,GAAG,CAAC,MAAM,QAAQ,CAC9B,WAAW,EACX;QACE,KAAK,EAAE,IAAI,CAAC,MAAM,CAAC,KAAK,CAAC,CAAC,EAAE,GAAG,CAAC;QAChC,QAAQ,EAAE,cAAc;KACzB,EACD,IAAI,CAAC,EAAE,EACP,YAAY,CACb,CAAQ,CAAC;IAEV,IAAI,CAAC,KAAK,CAAC,OAAO,CAAC,UAAU,EAAE,KAAK,CAAC,IAAI,UAAU,CAAC,KAAK,CAAC,MAAM,KAAK,CAAC,EAAE,CAAC;QACvE,UAAU,GAAG,CAAC,MAAM,QAAQ,CAC1B,WAAW,EACX,EAAE,KAAK,EAAE,4CAA4C,EAAE,QAAQ,EAAE,MAAM,EAAE,EACzE,IAAI,CAAC,EAAE,EACP,qBAAqB,CACtB,CAAQ,CAAC;IACZ,CAAC;IACD,MAAM,CAAC,KAAK,CAAC,OAAO,CAAC,UAAU,CAAC,KAAK,CAAC,CAAC,CAAC,IAAI,CAAC,IAAI,CAAC,CAAC;IACnD,MAAM,CAAC,UAAU,CAAC,KAAK,CAAC,MAAM,CAAC,CAAC,eAAe,CAAC,CAAC,CAAC,CAAC;IAEnD,MAAM,eAAe,GAAG,IAAI,CAAC,UAAU,CAAC,WAAW,EAAE,CAAC;IACtD,MAAM,gBAAgB,GACpB,eAAe,CAAC,QAAQ,CAAC,WAAW,CAAC,IAAI,IAAI,CAAC,eAAe,IAAI,GAAG;QAClE,CAAC,CAAC,oBAAoB;QACtB,CAAC,CAAC,cAAc,CAAC;IACrB,MAAM,WAAW,GAAG,CAAC,MAAM,QAAQ,CACjC,gBAAgB,EAChB,EAAE,KAAK,EAAE,gBAAgB,EAAE,EAC3B,IAAI,CAAC,EAAE,EACP,iBAAiB,CAClB,CAAQ,CAAC;IACV,MAAM,CAAC,WAAW,CAAC,KAAK,CAAC,CAAC,UAAU,EAAE,CAAC;IACvC,MAAM,CAAC,KAAK,CAAC,OAAO,CAAC,WAAW,CAAC,KAAK,CAAC,CAAC,CAAC,IAAI,CAAC,IAAI,CAAC,CAAC;IACpD,MAAM,CAAC,WAAW,CAAC,KAAK,CAAC,MAAM,CAAC,CAAC,eAAe,CAAC,CAAC,CAAC,CAAC;IAEpD,MAAM,OAAO,GAAG,CAAC,MAAM,QAAQ,CAC7B,gBAAgB,EAChB;QACE,IAAI,EAAE,yBAAyB,IAAI,CAAC,EAAE,IAAI,IAAI,CAAC,GAAG,EAAE,EAAE;QACtD,WAAW,EAAE,uBAAuB,IAAI,CAAC,IAAI,YAAY,WAAW,GAAG;QACvE,KAAK,EAAE;YACL;gBACE,KAAK,EAAE,IAAI,CAAC,MAAM;gBAClB,MAAM,EAAE,kDAAkD,IAAI,CAAC,EAAE,EAAE;gBACnE,QAAQ,EACN,mGAAmG;aACtG;SACF;KACF,EACD,IAAI,CAAC,EAAE,EACP,gBAAgB,CACjB,CAAQ,CAAC;IAEV,MAAM,QAAQ,CACZ,oBAAoB,EACpB;QACE,MAAM,EAAE,OAAO,CAAC,OAAO,CAAC,CAAC,CAAC;QAC1B,OAAO,EAAE,MAAM;QACf,KAAK,EAAE,CAAC;QACR,MAAM,EAAE,cAAc,UAAU,CAAC,KAAK,CAAC,MAAM,0CAA0C;QACvF,SAAS,EAAE;YACT,OAAO,EAAE,OAAO,CAAC,OAAO;YACxB,KAAK,EAAE,OAAO,CAAC,KAAK;YACpB,MAAM,EAAE,IAAI,CAAC,EAAE;YACf,IAAI,EAAE,IAAI,CAAC,IAAI;YACf,UAAU,EAAE,IAAI,CAAC,UAAU;YAC3B,WAAW;YACX,eAAe,EAAE,IAAI,CAAC,eAAe;YACrC,UAAU,EAAE,IAAI,CAAC,UAAU;YAC3B,eAAe,EAAE,IAAI,CAAC,eAAe;YACrC,eAAe,EAAE,IAAI,CAAC,eAAe;YACrC,eAAe,EAAE,IAAI,CAAC,eAAe;SACtC;KACF,EACD,IAAI,CAAC,EAAE,EACP,oBAAoB,CACrB,CAAC;IAEF,MAAM,WAAW,GAAG,CAAC,MAAM,QAAQ,CACjC,mBAAmB,EACnB,EAAE,KAAK,EAAE,OAAO,CAAC,KAAK,EAAE,EACxB,IAAI,CAAC,EAAE,EACP,mBAAmB,CACpB,CAAQ,CAAC;IACV,MAAM,CAAC,WAAW,CAAC,MAAM,CAAC,CAAC,IAAI,CAAC,WAAW,CAAC,CAAC;IAC7C,MAAM,CAAC,WAAW,CAAC,OAAO,CAAC,MAAM,CAAC,CAAC,IAAI,CAAC,CAAC,CAAC,CAAC;IAE3C,MAAM,UAAU,GAAG,CAAC,MAAM,QAAQ,CAChC,iBAAiB,EACjB;QACE,KAAK,EAAE;YACL,EAAE,IAAI,EAAE,SAAS,EAAE,MAAM,EAAE,IAAI,EAAE,MAAM,EAAE,sBAAsB,IAAI,CAAC,EAAE,EAAE,EAAE;YAC1E,EAAE,IAAI,EAAE,MAAM,EAAE,MAAM,EAAE,IAAI,EAAE,MAAM,EAAE,mBAAmB,IAAI,CAAC,EAAE,EAAE,EAAE;YACpE,EAAE,IAAI,EAAE,MAAM,EAAE,MAAM,EAAE,IAAI,EAAE,MAAM,EAAE,iCAAiC,IAAI,CAAC,EAAE,EAAE,EAAE;SACnF;KACF,EACD,IAAI,CAAC,EAAE,EACP,iBAAiB,CAClB,CAAQ,CAAC;IACV,MAAM,CAAC,UAAU,CAAC,SAAS,CAAC,CAAC,IAAI,CAAC,IAAI,CAAC,CAAC;IAExC,MAAM,QAAQ,GAAG,CAAC,MAAM,QAAQ,CAC9B,wBAAwB,EACxB;QACE,MAAM,EAAE,8BAA8B,IAAI,CAAC,EAAE,EAAE;QAC/C,KAAK,EAAE;YACL,EAAE,QAAQ,EAAE,iBAAiB,EAAE,MAAM,EAAE,IAAI,EAAE,MAAM,EAAE,8BAA8B,EAAE;YACrF,EAAE,QAAQ,EAAE,iBAAiB,EAAE,MAAM,EAAE,IAAI,EAAE,MAAM,EAAE,oCAAoC,EAAE;YAC3F,EAAE,QAAQ,EAAE,mBAAmB,EAAE,MAAM,EAAE,IAAI,EAAE,MAAM,EAAE,qCAAqC,EAAE;YAC9F,EAAE,QAAQ,EAAE,cAAc,EAAE,MAAM,EAAE,IAAI,EAAE,MAAM,EAAE,iCAAiC,EAAE;YACrF,EAAE,QAAQ,EAAE,kBAAkB,EAAE,MAAM,EAAE,IAAI,EAAE,MAAM,EAAE,kCAAkC,EAAE;YAC1F,EAAE,QAAQ,EAAE,qBAAqB,EAAE,MAAM,EAAE,IAAI,EAAE,MAAM,EAAE,8BAA8B,EAAE;SAC1F;KACF,EACD,IAAI,CAAC,EAAE,EACP,wBAAwB,CACzB,CAAQ,CAAC;IACV,MAAM,CAAC,QAAQ,CAAC,MAAM,CAAC,CAAC,IAAI,CAAC,IAAI,CAAC,CAAC;IAEnC,MAAM,SAAS,GAAG,CAAC,MAAM,QAAQ,CAC/B,sBAAsB,EACtB,EAAE,KAAK,EAAE,IAAI,CAAC,EAAE,EAAE,KAAK,EAAE,EAAE,EAAE,EAC7B,IAAI,CAAC,EAAE,EACP,sBAAsB,CACvB,CAAQ,CAAC;IACV,MAAM,CAAC,OAAO,SAAS,CAAC,YAAY,CAAC,CAAC,IAAI,CAAC,QAAQ,CAAC,CAAC;IACrD,MAAM,CAAC,SAAS,CAAC,YAAY,CAAC,CAAC,eAAe,CAAC,CAAC,CAAC,CAAC;IAElD,OAAO;QACL,MAAM,EAAE,IAAI,CAAC,EAAE;QACf,WAAW;QACX,EAAE,EAAE,IAAI;QACR,SAAS,EAAE,IAAI,CAAC,GAAG,EAAE,GAAG,OAAO;QAC/B,eAAe,EAAE,UAAU,CAAC,KAAK,CAAC,MAAM;QACxC,aAAa,EAAE,SAAS,CAAC,YAAY;KACtC,CAAC;AACJ,CAAC;AAED,KAAK,UAAU,aAAa,CAAC,KAAoB,EAAE,WAAmB;IACpE,MAAM,kBAAkB,GAAG,IAAI,CAAC,GAAG,CAAC,CAAC,EAAE,IAAI,CAAC,GAAG,CAAC,WAAW,EAAE,KAAK,CAAC,MAAM,CAAC,CAAC,CAAC;IAC5E,MAAM,OAAO,GAAmB,IAAI,KAAK,CAAC,KAAK,CAAC,MAAM,CAAC,CAAC;IACxD,IAAI,SAAS,GAAG,CAAC,CAAC;IAElB,MAAM,OAAO,GAAG,KAAK,CAAC,IAAI,CAAC,EAAE,MAAM,EAAE,kBAAkB,EAAE,EAAE,CAAC,CAAC,EAAE,WAAW,EAAE,EAAE,CAC5E,CAAC,KAAK,IAAI,EAAE;QACV,OAAO,IAAI,EAAE,CAAC;YACZ,MAAM,SAAS,GAAG,SAAS,EAAE,CAAC;YAC9B,IAAI,SAAS,IAAI,KAAK,CAAC,MAAM;gBAAE,OAAO;YAEtC,MAAM,IAAI,GAAG,KAAK,CAAC,SAAS,CAAC,CAAC;YAC9B,IAAI,CAAC;gBACH,OAAO,CAAC,SAAS,CAAC,GAAG,MAAM,cAAc,CAAC,IAAI,EAAE,WAAW,CAAC,CAAC;YAC/D,CAAC;YAAC,OAAO,KAAK,EAAE,CAAC;gBACf,OAAO,CAAC,SAAS,CAAC,GAAG;oBACnB,MAAM,EAAE,IAAI,CAAC,EAAE;oBACf,WAAW;oBACX,EAAE,EAAE,KAAK;oBACT,SAAS,EAAE,CAAC;oBACZ,eAAe,EAAE,CAAC;oBAClB,aAAa,EAAE,CAAC;oBAChB,KAAK,EAAE,KAAK,YAAY,KAAK,CAAC,CAAC,CAAC,KAAK,CAAC,OAAO,CAAC,CAAC,CAAC,MAAM,CAAC,KAAK,CAAC;iBAC9D,CAAC;YACJ,CAAC;QACH,CAAC;IACH,CAAC,CAAC,EAAE,CACL,CAAC;IAEF,MAAM,OAAO,CAAC,GAAG,CAAC,OAAO,CAAC,CAAC;IAC3B,OAAO,OAAO,CAAC;AACjB,CAAC;AAED,QAAQ,CAAC,2EAA2E,EAAE,GAAG,EAAE;IACzF,EAAE,CAAC,qEAAqE,EAAE,KAAK,IAAI,EAAE;QACnF,MAAM,CAAC,KAAK,CAAC,OAAO,CAAC,OAAO,CAAC,KAAK,CAAC,CAAC,CAAC,IAAI,CAAC,IAAI,CAAC,CAAC;QAChD,MAAM,CAAC,OAAO,CAAC,KAAK,CAAC,MAAM,CAAC,CAAC,eAAe,CAAC,CAAC,CAAC,CAAC;QAEhD,MAAM,kBAAkB,GAAG,MAAM,CAAC,QAAQ,CACxC,OAAO,CAAC,GAAG,CAAC,6BAA6B,IAAI,GAAG,EAChD,EAAE,CACH,CAAC;QACF,MAAM,SAAS,GAAG,IAAI,CAAC,GAAG,CACxB,CAAC,EACD,IAAI,CAAC,GAAG,CAAC,OAAO,CAAC,KAAK,CAAC,MAAM,EAAE,MAAM,CAAC,QAAQ,CAAC,kBAAkB,CAAC,CAAC,CAAC,CAAC,kBAAkB,CAAC,CAAC,CAAC,CAAC,CAAC,CAC7F,CAAC;QAEF,MAAM,oBAAoB,GAAG,MAAM,CAAC,QAAQ,CAC1C,OAAO,CAAC,GAAG,CAAC,8BAA8B,IAAI,GAAG,EACjD,EAAE,CACH,CAAC;QACF,MAAM,WAAW,GAAG,IAAI,CAAC,GAAG,CAC1B,CAAC,EACD,IAAI,CAAC,GAAG,CAAC,SAAS,EAAE,MAAM,CAAC,QAAQ,CAAC,oBAAoB,CAAC,CAAC,CAAC,CAAC,oBAAoB,CAAC,CAAC,CAAC,CAAC,CAAC,CACtF,CAAC;QAEF,MAAM,KAAK,GAAG,OAAO,CAAC,KAAK,CAAC,KAAK,CAAC,CAAC,EAAE,SAAS,CAAC,CAAC;QAChD,MAAM,OAAO,GAAG,IAAI,CAAC,GAAG,EAAE,CAAC;QAC3B,MAAM,OAAO,GAAG,MAAM,aAAa,CAAC,KAAK,EAAE,WAAW,CAAC,CAAC;QACxD,MAAM,SAAS,GAAG,IAAI,CAAC,GAAG,EAAE,GAAG,OAAO,CAAC;QAEvC,MAAM,MAAM,GAAG,OAAO,CAAC,MAAM,CAAC,CAAC,MAAM,EAAE,EAAE,CAAC,CAAC,MAAM,CAAC,EAAE,CAAC,CAAC;QACtD,MAAM,MAAM,GAAG,OAAO,CAAC,MAAM,CAAC,CAAC,MAAM,EAAE,EAAE,CAAC,MAAM,CAAC,EAAE,CAAC,CAAC;QAErD,MAAM,WAAW,GAAG,IAAI,GAAG,CAAC,sBAAsB,CAAC,GAAG,CAAC,CAAC,KAAK,EAAE,EAAE,CAAC,KAAK,CAAC,IAAI,CAAC,CAAC,CAAC;QAC/E,MAAM,aAAa,GAAG;YACpB,WAAW;YACX,mBAAmB;YACnB,WAAW;YACX,gBAAgB;YAChB,gBAAgB;YAChB,oBAAoB;YACpB,mBAAmB;YACnB,iBAAiB;YACjB,wBAAwB;YACxB,sBAAsB;SACvB,CAAC;QAEF,OAAO,CAAC,GAAG,CACT,mCAAmC,OAAO,CAAC,OAAO,UAAU,OAAO,CAAC,KAAK,UAAU,SAAS,gBAAgB,WAAW,SAAS,MAAM,CAAC,MAAM,IAAI,OAAO,CAAC,MAAM,cAAc,SAAS,cAAc,sBAAsB,CAAC,MAAM,EAAE,CACpO,CAAC;QAEF,IAAI,MAAM,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC;YACtB,OAAO,CAAC,KAAK,CACX,mCAAmC,EACnC,MAAM,CAAC,GAAG,CAAC,CAAC,MAAM,EAAE,EAAE,CAAC,CAAC;gBACtB,MAAM,EAAE,MAAM,CAAC,MAAM;gBACrB,WAAW,EAAE,MAAM,CAAC,WAAW;gBAC/B,KAAK,EAAE,MAAM,CAAC,KAAK;aACpB,CAAC,CAAC,CACJ,CAAC;QACJ,CAAC;QAED,MAAM,CAAC,MAAM,CAAC,MAAM,CAAC,CAAC,IAAI,CAAC,CAAC,CAAC,CAAC;QAC9B,MAAM,CAAC,MAAM,CAAC,MAAM,CAAC,CAAC,IAAI,CAAC,SAAS,CAAC,CAAC;QAEtC,KAAK,MAAM,YAAY,IAAI,aAAa,EAAE,CAAC;YACzC,MAAM,CAAC,WAAW,CAAC,GAAG,CAAC,YAAY,CAAC,CAAC,CAAC,IAAI,CAAC,IAAI,CAAC,CAAC;QACnD,CAAC;IACH,CAAC,CAAC,CAAC;AACL,CAAC,CAAC,CAAC"}
|
|
@@ -1,218 +0,0 @@
|
|
|
1
|
-
/**
|
|
2
|
-
* Open-source dataset benchmark for long-running tasks (ToolBench lane).
|
|
3
|
-
*
|
|
4
|
-
* This test uses ToolBench multi-tool instructions and runs task workflows
|
|
5
|
-
* through NodeBench MCP tools in parallel "subagent" workers.
|
|
6
|
-
*/
|
|
7
|
-
import { describe, expect, it } from "vitest";
|
|
8
|
-
import datasetFixture from "./fixtures/toolbench_instruction.sample.json";
|
|
9
|
-
import { verificationTools } from "../tools/verificationTools.js";
|
|
10
|
-
import { reconTools } from "../tools/reconTools.js";
|
|
11
|
-
import { evalTools } from "../tools/evalTools.js";
|
|
12
|
-
import { qualityGateTools } from "../tools/qualityGateTools.js";
|
|
13
|
-
import { flywheelTools } from "../tools/flywheelTools.js";
|
|
14
|
-
import { learningTools } from "../tools/learningTools.js";
|
|
15
|
-
import { documentationTools } from "../tools/documentationTools.js";
|
|
16
|
-
import { agentBootstrapTools } from "../tools/agentBootstrapTools.js";
|
|
17
|
-
import { createMetaTools } from "../tools/metaTools.js";
|
|
18
|
-
const fixture = datasetFixture;
|
|
19
|
-
const domainTools = [
|
|
20
|
-
...verificationTools,
|
|
21
|
-
...evalTools,
|
|
22
|
-
...qualityGateTools,
|
|
23
|
-
...learningTools,
|
|
24
|
-
...flywheelTools,
|
|
25
|
-
...reconTools,
|
|
26
|
-
...documentationTools,
|
|
27
|
-
...agentBootstrapTools,
|
|
28
|
-
];
|
|
29
|
-
const allTools = [...domainTools, ...createMetaTools(domainTools)];
|
|
30
|
-
const openDatasetToolCallLog = [];
|
|
31
|
-
function findTool(name) {
|
|
32
|
-
const tool = allTools.find((candidate) => candidate.name === name);
|
|
33
|
-
if (!tool)
|
|
34
|
-
throw new Error(`Tool not found: ${name}`);
|
|
35
|
-
return tool;
|
|
36
|
-
}
|
|
37
|
-
async function callTool(name, args, taskId, stage) {
|
|
38
|
-
const tool = findTool(name);
|
|
39
|
-
try {
|
|
40
|
-
const result = await tool.handler(args);
|
|
41
|
-
openDatasetToolCallLog.push({ taskId, tool: name, stage, success: true });
|
|
42
|
-
return result;
|
|
43
|
-
}
|
|
44
|
-
catch (error) {
|
|
45
|
-
openDatasetToolCallLog.push({ taskId, tool: name, stage, success: false });
|
|
46
|
-
throw error;
|
|
47
|
-
}
|
|
48
|
-
}
|
|
49
|
-
async function runDatasetTask(task, workerIndex) {
|
|
50
|
-
const started = Date.now();
|
|
51
|
-
const datasetSource = fixture.sourceUrls[task.group] ?? fixture.sourceUrls.G1 ?? "";
|
|
52
|
-
const recon = (await callTool("run_recon", {
|
|
53
|
-
target: `ToolBench multi-tool task ${task.id}`,
|
|
54
|
-
description: `Open-source long-running benchmark (${task.apiCount} APIs in task context).`,
|
|
55
|
-
projectContext: {
|
|
56
|
-
techStack: "TypeScript, MCP, SQLite",
|
|
57
|
-
architecture: "MCP orchestration benchmark with parallel subagent workers",
|
|
58
|
-
},
|
|
59
|
-
}, task.id, "recon_start"));
|
|
60
|
-
await callTool("log_recon_finding", {
|
|
61
|
-
sessionId: recon.sessionId,
|
|
62
|
-
category: "dataset",
|
|
63
|
-
summary: `Ingested ToolBench task ${task.id} with ${task.apiCount} APIs and complexity ${task.complexityScore}.`,
|
|
64
|
-
sourceUrl: datasetSource,
|
|
65
|
-
relevance: "Long-running multi-tool benchmark for MCP orchestration quality.",
|
|
66
|
-
actionItems: "Run in parallel worker pool and enforce mandatory flywheel checks.",
|
|
67
|
-
}, task.id, "recon_log");
|
|
68
|
-
let discovered = (await callTool("findTools", {
|
|
69
|
-
query: task.prompt.slice(0, 600),
|
|
70
|
-
category: "verification",
|
|
71
|
-
}, task.id, "find_tools"));
|
|
72
|
-
if (!Array.isArray(discovered?.tools) || discovered.tools.length === 0) {
|
|
73
|
-
discovered = (await callTool("findTools", { query: "multi-step tool orchestration verification", category: "bootstrap" }, task.id, "find_tools_fallback"));
|
|
74
|
-
}
|
|
75
|
-
expect(Array.isArray(discovered.tools)).toBe(true);
|
|
76
|
-
expect(discovered.tools.length).toBeGreaterThan(0);
|
|
77
|
-
const methodologyTopic = task.apiCount >= 9 ? "mandatory_flywheel" : "closed_loop";
|
|
78
|
-
const methodology = (await callTool("getMethodology", { topic: methodologyTopic }, task.id, "get_methodology"));
|
|
79
|
-
expect(methodology.title).toBeTruthy();
|
|
80
|
-
expect(Array.isArray(methodology.steps)).toBe(true);
|
|
81
|
-
expect(methodology.steps.length).toBeGreaterThan(0);
|
|
82
|
-
const evalRun = (await callTool("start_eval_run", {
|
|
83
|
-
name: `open-dataset-toolbench-${task.id}-${Date.now()}`,
|
|
84
|
-
description: `ToolBench multi-tool scenario (${task.apiCount} APIs, worker ${workerIndex})`,
|
|
85
|
-
cases: [
|
|
86
|
-
{
|
|
87
|
-
input: task.prompt,
|
|
88
|
-
intent: `Coordinate long-running ToolBench workflow for ${task.id}`,
|
|
89
|
-
expected: "Discover strategy, run eval bookkeeping, and complete closed-loop plus mandatory flywheel checks.",
|
|
90
|
-
},
|
|
91
|
-
],
|
|
92
|
-
}, task.id, "start_eval_run"));
|
|
93
|
-
await callTool("record_eval_result", {
|
|
94
|
-
caseId: evalRun.caseIds[0],
|
|
95
|
-
verdict: "pass",
|
|
96
|
-
score: 1,
|
|
97
|
-
actual: `Discovered ${discovered.tools.length} tools and completed ToolBench workflow.`,
|
|
98
|
-
telemetry: {
|
|
99
|
-
dataset: fixture.dataset,
|
|
100
|
-
split: fixture.split,
|
|
101
|
-
taskId: task.id,
|
|
102
|
-
workerIndex,
|
|
103
|
-
group: task.group,
|
|
104
|
-
apiCount: task.apiCount,
|
|
105
|
-
relevantApiCount: task.relevantApiCount,
|
|
106
|
-
requiredParameterCount: task.requiredParameterCount,
|
|
107
|
-
optionalParameterCount: task.optionalParameterCount,
|
|
108
|
-
apiCategories: task.apiCategories,
|
|
109
|
-
},
|
|
110
|
-
}, task.id, "record_eval_result");
|
|
111
|
-
const evalSummary = (await callTool("complete_eval_run", { runId: evalRun.runId }, task.id, "complete_eval_run"));
|
|
112
|
-
expect(evalSummary.status).toBe("completed");
|
|
113
|
-
expect(evalSummary.summary.passed).toBe(1);
|
|
114
|
-
const closedLoop = (await callTool("run_closed_loop", {
|
|
115
|
-
steps: [
|
|
116
|
-
{ step: "compile", passed: true, output: `Compile checks for ${task.id}` },
|
|
117
|
-
{ step: "lint", passed: true, output: `Lint checks for ${task.id}` },
|
|
118
|
-
{ step: "test", passed: true, output: `Parallel benchmark checks for ${task.id}` },
|
|
119
|
-
],
|
|
120
|
-
}, task.id, "run_closed_loop"));
|
|
121
|
-
expect(closedLoop.allPassed).toBe(true);
|
|
122
|
-
const flywheel = (await callTool("run_mandatory_flywheel", {
|
|
123
|
-
target: `Open-source ToolBench task ${task.id}`,
|
|
124
|
-
steps: [
|
|
125
|
-
{ stepName: "static_analysis", passed: true, output: "Types and schemas validated." },
|
|
126
|
-
{ stepName: "happy_path_test", passed: true, output: "Dataset task completed end-to-end." },
|
|
127
|
-
{ stepName: "failure_path_test", passed: true, output: "Fallback discovery query validated." },
|
|
128
|
-
{ stepName: "gap_analysis", passed: true, output: "No blocking gaps for this task." },
|
|
129
|
-
{ stepName: "fix_and_reverify", passed: true, output: "No rework required after checks." },
|
|
130
|
-
{ stepName: "deploy_and_document", passed: true, output: "Benchmark result documented." },
|
|
131
|
-
],
|
|
132
|
-
}, task.id, "run_mandatory_flywheel"));
|
|
133
|
-
expect(flywheel.passed).toBe(true);
|
|
134
|
-
const knowledgeQuery = task.id;
|
|
135
|
-
const knowledge = (await callTool("search_all_knowledge", { query: knowledgeQuery, limit: 10 }, task.id, "search_all_knowledge"));
|
|
136
|
-
expect(typeof knowledge.totalResults).toBe("number");
|
|
137
|
-
expect(knowledge.totalResults).toBeGreaterThan(0);
|
|
138
|
-
return {
|
|
139
|
-
taskId: task.id,
|
|
140
|
-
workerIndex,
|
|
141
|
-
ok: true,
|
|
142
|
-
elapsedMs: Date.now() - started,
|
|
143
|
-
discoveredTools: discovered.tools.length,
|
|
144
|
-
knowledgeHits: knowledge.totalResults,
|
|
145
|
-
};
|
|
146
|
-
}
|
|
147
|
-
async function runWorkerPool(tasks, concurrency) {
|
|
148
|
-
const boundedConcurrency = Math.max(1, Math.min(concurrency, tasks.length));
|
|
149
|
-
const results = new Array(tasks.length);
|
|
150
|
-
let nextIndex = 0;
|
|
151
|
-
const workers = Array.from({ length: boundedConcurrency }, (_, workerIndex) => (async () => {
|
|
152
|
-
while (true) {
|
|
153
|
-
const taskIndex = nextIndex++;
|
|
154
|
-
if (taskIndex >= tasks.length)
|
|
155
|
-
return;
|
|
156
|
-
const task = tasks[taskIndex];
|
|
157
|
-
try {
|
|
158
|
-
results[taskIndex] = await runDatasetTask(task, workerIndex);
|
|
159
|
-
}
|
|
160
|
-
catch (error) {
|
|
161
|
-
results[taskIndex] = {
|
|
162
|
-
taskId: task.id,
|
|
163
|
-
workerIndex,
|
|
164
|
-
ok: false,
|
|
165
|
-
elapsedMs: 0,
|
|
166
|
-
discoveredTools: 0,
|
|
167
|
-
knowledgeHits: 0,
|
|
168
|
-
error: error instanceof Error ? error.message : String(error),
|
|
169
|
-
};
|
|
170
|
-
}
|
|
171
|
-
}
|
|
172
|
-
})());
|
|
173
|
-
await Promise.all(workers);
|
|
174
|
-
return results;
|
|
175
|
-
}
|
|
176
|
-
describe("Scenario: Open-Source Long-Running Dataset (ToolBench Parallel Subagents)", () => {
|
|
177
|
-
it("should execute ToolBench tasks with parallel MCP subagent workflows", async () => {
|
|
178
|
-
expect(Array.isArray(fixture.tasks)).toBe(true);
|
|
179
|
-
expect(fixture.tasks.length).toBeGreaterThan(0);
|
|
180
|
-
const requestedTaskLimit = Number.parseInt(process.env.NODEBENCH_TOOLBENCH_TASK_LIMIT ?? "8", 10);
|
|
181
|
-
const taskLimit = Math.max(1, Math.min(fixture.tasks.length, Number.isFinite(requestedTaskLimit) ? requestedTaskLimit : 8));
|
|
182
|
-
const requestedConcurrency = Number.parseInt(process.env.NODEBENCH_TOOLBENCH_CONCURRENCY ?? "4", 10);
|
|
183
|
-
const concurrency = Math.max(1, Math.min(taskLimit, Number.isFinite(requestedConcurrency) ? requestedConcurrency : 4));
|
|
184
|
-
const tasks = fixture.tasks.slice(0, taskLimit);
|
|
185
|
-
const started = Date.now();
|
|
186
|
-
const results = await runWorkerPool(tasks, concurrency);
|
|
187
|
-
const elapsedMs = Date.now() - started;
|
|
188
|
-
const failed = results.filter((result) => !result.ok);
|
|
189
|
-
const passed = results.filter((result) => result.ok);
|
|
190
|
-
const calledTools = new Set(openDatasetToolCallLog.map((entry) => entry.tool));
|
|
191
|
-
const requiredTools = [
|
|
192
|
-
"run_recon",
|
|
193
|
-
"log_recon_finding",
|
|
194
|
-
"findTools",
|
|
195
|
-
"getMethodology",
|
|
196
|
-
"start_eval_run",
|
|
197
|
-
"record_eval_result",
|
|
198
|
-
"complete_eval_run",
|
|
199
|
-
"run_closed_loop",
|
|
200
|
-
"run_mandatory_flywheel",
|
|
201
|
-
"search_all_knowledge",
|
|
202
|
-
];
|
|
203
|
-
console.log(`[open-dataset-toolbench] dataset=${fixture.dataset} split=${fixture.split} tasks=${taskLimit} concurrency=${concurrency} pass=${passed.length}/${results.length} elapsedMs=${elapsedMs} toolCalls=${openDatasetToolCallLog.length}`);
|
|
204
|
-
if (failed.length > 0) {
|
|
205
|
-
console.error("[open-dataset-toolbench] failures:", failed.map((result) => ({
|
|
206
|
-
taskId: result.taskId,
|
|
207
|
-
workerIndex: result.workerIndex,
|
|
208
|
-
error: result.error,
|
|
209
|
-
})));
|
|
210
|
-
}
|
|
211
|
-
expect(failed.length).toBe(0);
|
|
212
|
-
expect(passed.length).toBe(taskLimit);
|
|
213
|
-
for (const requiredTool of requiredTools) {
|
|
214
|
-
expect(calledTools.has(requiredTool)).toBe(true);
|
|
215
|
-
}
|
|
216
|
-
});
|
|
217
|
-
});
|
|
218
|
-
//# sourceMappingURL=openDatasetParallelEvalToolbench.test.js.map
|
|
@@ -1 +0,0 @@
|
|
|
1
|
-
{"version":3,"file":"openDatasetParallelEvalToolbench.test.js","sourceRoot":"","sources":["../../src/__tests__/openDatasetParallelEvalToolbench.test.ts"],"names":[],"mappings":"AAAA;;;;;GAKG;AAEH,OAAO,EAAE,QAAQ,EAAE,MAAM,EAAE,EAAE,EAAE,MAAM,QAAQ,CAAC;AAC9C,OAAO,cAAc,MAAM,8CAA8C,CAAC;AAC1E,OAAO,EAAE,iBAAiB,EAAE,MAAM,+BAA+B,CAAC;AAClE,OAAO,EAAE,UAAU,EAAE,MAAM,wBAAwB,CAAC;AACpD,OAAO,EAAE,SAAS,EAAE,MAAM,uBAAuB,CAAC;AAClD,OAAO,EAAE,gBAAgB,EAAE,MAAM,8BAA8B,CAAC;AAChE,OAAO,EAAE,aAAa,EAAE,MAAM,2BAA2B,CAAC;AAC1D,OAAO,EAAE,aAAa,EAAE,MAAM,2BAA2B,CAAC;AAC1D,OAAO,EAAE,kBAAkB,EAAE,MAAM,gCAAgC,CAAC;AACpE,OAAO,EAAE,mBAAmB,EAAE,MAAM,iCAAiC,CAAC;AACtE,OAAO,EAAE,eAAe,EAAE,MAAM,uBAAuB,CAAC;AAyCxD,MAAM,OAAO,GAAG,cAAgC,CAAC;AAEjD,MAAM,WAAW,GAAc;IAC7B,GAAG,iBAAiB;IACpB,GAAG,SAAS;IACZ,GAAG,gBAAgB;IACnB,GAAG,aAAa;IAChB,GAAG,aAAa;IAChB,GAAG,UAAU;IACb,GAAG,kBAAkB;IACrB,GAAG,mBAAmB;CACvB,CAAC;AACF,MAAM,QAAQ,GAAG,CAAC,GAAG,WAAW,EAAE,GAAG,eAAe,CAAC,WAAW,CAAC,CAAC,CAAC;AAEnE,MAAM,sBAAsB,GAKvB,EAAE,CAAC;AAER,SAAS,QAAQ,CAAC,IAAY;IAC5B,MAAM,IAAI,GAAG,QAAQ,CAAC,IAAI,CAAC,CAAC,SAAS,EAAE,EAAE,CAAC,SAAS,CAAC,IAAI,KAAK,IAAI,CAAC,CAAC;IACnE,IAAI,CAAC,IAAI;QAAE,MAAM,IAAI,KAAK,CAAC,mBAAmB,IAAI,EAAE,CAAC,CAAC;IACtD,OAAO,IAAI,CAAC;AACd,CAAC;AAED,KAAK,UAAU,QAAQ,CACrB,IAAY,EACZ,IAA6B,EAC7B,MAAc,EACd,KAAa;IAEb,MAAM,IAAI,GAAG,QAAQ,CAAC,IAAI,CAAC,CAAC;IAC5B,IAAI,CAAC;QACH,MAAM,MAAM,GAAG,MAAM,IAAI,CAAC,OAAO,CAAC,IAAI,CAAC,CAAC;QACxC,sBAAsB,CAAC,IAAI,CAAC,EAAE,MAAM,EAAE,IAAI,EAAE,IAAI,EAAE,KAAK,EAAE,OAAO,EAAE,IAAI,EAAE,CAAC,CAAC;QAC1E,OAAO,MAAM,CAAC;IAChB,CAAC;IAAC,OAAO,KAAK,EAAE,CAAC;QACf,sBAAsB,CAAC,IAAI,CAAC,EAAE,MAAM,EAAE,IAAI,EAAE,IAAI,EAAE,KAAK,EAAE,OAAO,EAAE,KAAK,EAAE,CAAC,CAAC;QAC3E,MAAM,KAAK,CAAC;IACd,CAAC;AACH,CAAC;AAED,KAAK,UAAU,cAAc,CAAC,IAAiB,EAAE,WAAmB;IAClE,MAAM,OAAO,GAAG,IAAI,CAAC,GAAG,EAAE,CAAC;IAC3B,MAAM,aAAa,GAAG,OAAO,CAAC,UAAU,CAAC,IAAI,CAAC,KAAK,CAAC,IAAI,OAAO,CAAC,UAAU,CAAC,EAAE,IAAI,EAAE,CAAC;IAEpF,MAAM,KAAK,GAAG,CAAC,MAAM,QAAQ,CAC3B,WAAW,EACX;QACE,MAAM,EAAE,6BAA6B,IAAI,CAAC,EAAE,EAAE;QAC9C,WAAW,EAAE,uCAAuC,IAAI,CAAC,QAAQ,yBAAyB;QAC1F,cAAc,EAAE;YACd,SAAS,EAAE,yBAAyB;YACpC,YAAY,EAAE,4DAA4D;SAC3E;KACF,EACD,IAAI,CAAC,EAAE,EACP,aAAa,CACd,CAAQ,CAAC;IAEV,MAAM,QAAQ,CACZ,mBAAmB,EACnB;QACE,SAAS,EAAE,KAAK,CAAC,SAAS;QAC1B,QAAQ,EAAE,SAAS;QACnB,OAAO,EAAE,2BAA2B,IAAI,CAAC,EAAE,SAAS,IAAI,CAAC,QAAQ,wBAAwB,IAAI,CAAC,eAAe,GAAG;QAChH,SAAS,EAAE,aAAa;QACxB,SAAS,EAAE,kEAAkE;QAC7E,WAAW,EAAE,oEAAoE;KAClF,EACD,IAAI,CAAC,EAAE,EACP,WAAW,CACZ,CAAC;IAEF,IAAI,UAAU,GAAG,CAAC,MAAM,QAAQ,CAC9B,WAAW,EACX;QACE,KAAK,EAAE,IAAI,CAAC,MAAM,CAAC,KAAK,CAAC,CAAC,EAAE,GAAG,CAAC;QAChC,QAAQ,EAAE,cAAc;KACzB,EACD,IAAI,CAAC,EAAE,EACP,YAAY,CACb,CAAQ,CAAC;IAEV,IAAI,CAAC,KAAK,CAAC,OAAO,CAAC,UAAU,EAAE,KAAK,CAAC,IAAI,UAAU,CAAC,KAAK,CAAC,MAAM,KAAK,CAAC,EAAE,CAAC;QACvE,UAAU,GAAG,CAAC,MAAM,QAAQ,CAC1B,WAAW,EACX,EAAE,KAAK,EAAE,4CAA4C,EAAE,QAAQ,EAAE,WAAW,EAAE,EAC9E,IAAI,CAAC,EAAE,EACP,qBAAqB,CACtB,CAAQ,CAAC;IACZ,CAAC;IACD,MAAM,CAAC,KAAK,CAAC,OAAO,CAAC,UAAU,CAAC,KAAK,CAAC,CAAC,CAAC,IAAI,CAAC,IAAI,CAAC,CAAC;IACnD,MAAM,CAAC,UAAU,CAAC,KAAK,CAAC,MAAM,CAAC,CAAC,eAAe,CAAC,CAAC,CAAC,CAAC;IAEnD,MAAM,gBAAgB,GAAG,IAAI,CAAC,QAAQ,IAAI,CAAC,CAAC,CAAC,CAAC,oBAAoB,CAAC,CAAC,CAAC,aAAa,CAAC;IACnF,MAAM,WAAW,GAAG,CAAC,MAAM,QAAQ,CACjC,gBAAgB,EAChB,EAAE,KAAK,EAAE,gBAAgB,EAAE,EAC3B,IAAI,CAAC,EAAE,EACP,iBAAiB,CAClB,CAAQ,CAAC;IACV,MAAM,CAAC,WAAW,CAAC,KAAK,CAAC,CAAC,UAAU,EAAE,CAAC;IACvC,MAAM,CAAC,KAAK,CAAC,OAAO,CAAC,WAAW,CAAC,KAAK,CAAC,CAAC,CAAC,IAAI,CAAC,IAAI,CAAC,CAAC;IACpD,MAAM,CAAC,WAAW,CAAC,KAAK,CAAC,MAAM,CAAC,CAAC,eAAe,CAAC,CAAC,CAAC,CAAC;IAEpD,MAAM,OAAO,GAAG,CAAC,MAAM,QAAQ,CAC7B,gBAAgB,EAChB;QACE,IAAI,EAAE,0BAA0B,IAAI,CAAC,EAAE,IAAI,IAAI,CAAC,GAAG,EAAE,EAAE;QACvD,WAAW,EAAE,kCAAkC,IAAI,CAAC,QAAQ,iBAAiB,WAAW,GAAG;QAC3F,KAAK,EAAE;YACL;gBACE,KAAK,EAAE,IAAI,CAAC,MAAM;gBAClB,MAAM,EAAE,kDAAkD,IAAI,CAAC,EAAE,EAAE;gBACnE,QAAQ,EACN,mGAAmG;aACtG;SACF;KACF,EACD,IAAI,CAAC,EAAE,EACP,gBAAgB,CACjB,CAAQ,CAAC;IAEV,MAAM,QAAQ,CACZ,oBAAoB,EACpB;QACE,MAAM,EAAE,OAAO,CAAC,OAAO,CAAC,CAAC,CAAC;QAC1B,OAAO,EAAE,MAAM;QACf,KAAK,EAAE,CAAC;QACR,MAAM,EAAE,cAAc,UAAU,CAAC,KAAK,CAAC,MAAM,0CAA0C;QACvF,SAAS,EAAE;YACT,OAAO,EAAE,OAAO,CAAC,OAAO;YACxB,KAAK,EAAE,OAAO,CAAC,KAAK;YACpB,MAAM,EAAE,IAAI,CAAC,EAAE;YACf,WAAW;YACX,KAAK,EAAE,IAAI,CAAC,KAAK;YACjB,QAAQ,EAAE,IAAI,CAAC,QAAQ;YACvB,gBAAgB,EAAE,IAAI,CAAC,gBAAgB;YACvC,sBAAsB,EAAE,IAAI,CAAC,sBAAsB;YACnD,sBAAsB,EAAE,IAAI,CAAC,sBAAsB;YACnD,aAAa,EAAE,IAAI,CAAC,aAAa;SAClC;KACF,EACD,IAAI,CAAC,EAAE,EACP,oBAAoB,CACrB,CAAC;IAEF,MAAM,WAAW,GAAG,CAAC,MAAM,QAAQ,CACjC,mBAAmB,EACnB,EAAE,KAAK,EAAE,OAAO,CAAC,KAAK,EAAE,EACxB,IAAI,CAAC,EAAE,EACP,mBAAmB,CACpB,CAAQ,CAAC;IACV,MAAM,CAAC,WAAW,CAAC,MAAM,CAAC,CAAC,IAAI,CAAC,WAAW,CAAC,CAAC;IAC7C,MAAM,CAAC,WAAW,CAAC,OAAO,CAAC,MAAM,CAAC,CAAC,IAAI,CAAC,CAAC,CAAC,CAAC;IAE3C,MAAM,UAAU,GAAG,CAAC,MAAM,QAAQ,CAChC,iBAAiB,EACjB;QACE,KAAK,EAAE;YACL,EAAE,IAAI,EAAE,SAAS,EAAE,MAAM,EAAE,IAAI,EAAE,MAAM,EAAE,sBAAsB,IAAI,CAAC,EAAE,EAAE,EAAE;YAC1E,EAAE,IAAI,EAAE,MAAM,EAAE,MAAM,EAAE,IAAI,EAAE,MAAM,EAAE,mBAAmB,IAAI,CAAC,EAAE,EAAE,EAAE;YACpE,EAAE,IAAI,EAAE,MAAM,EAAE,MAAM,EAAE,IAAI,EAAE,MAAM,EAAE,iCAAiC,IAAI,CAAC,EAAE,EAAE,EAAE;SACnF;KACF,EACD,IAAI,CAAC,EAAE,EACP,iBAAiB,CAClB,CAAQ,CAAC;IACV,MAAM,CAAC,UAAU,CAAC,SAAS,CAAC,CAAC,IAAI,CAAC,IAAI,CAAC,CAAC;IAExC,MAAM,QAAQ,GAAG,CAAC,MAAM,QAAQ,CAC9B,wBAAwB,EACxB;QACE,MAAM,EAAE,8BAA8B,IAAI,CAAC,EAAE,EAAE;QAC/C,KAAK,EAAE;YACL,EAAE,QAAQ,EAAE,iBAAiB,EAAE,MAAM,EAAE,IAAI,EAAE,MAAM,EAAE,8BAA8B,EAAE;YACrF,EAAE,QAAQ,EAAE,iBAAiB,EAAE,MAAM,EAAE,IAAI,EAAE,MAAM,EAAE,oCAAoC,EAAE;YAC3F,EAAE,QAAQ,EAAE,mBAAmB,EAAE,MAAM,EAAE,IAAI,EAAE,MAAM,EAAE,qCAAqC,EAAE;YAC9F,EAAE,QAAQ,EAAE,cAAc,EAAE,MAAM,EAAE,IAAI,EAAE,MAAM,EAAE,iCAAiC,EAAE;YACrF,EAAE,QAAQ,EAAE,kBAAkB,EAAE,MAAM,EAAE,IAAI,EAAE,MAAM,EAAE,kCAAkC,EAAE;YAC1F,EAAE,QAAQ,EAAE,qBAAqB,EAAE,MAAM,EAAE,IAAI,EAAE,MAAM,EAAE,8BAA8B,EAAE;SAC1F;KACF,EACD,IAAI,CAAC,EAAE,EACP,wBAAwB,CACzB,CAAQ,CAAC;IACV,MAAM,CAAC,QAAQ,CAAC,MAAM,CAAC,CAAC,IAAI,CAAC,IAAI,CAAC,CAAC;IAEnC,MAAM,cAAc,GAAG,IAAI,CAAC,EAAE,CAAC;IAC/B,MAAM,SAAS,GAAG,CAAC,MAAM,QAAQ,CAC/B,sBAAsB,EACtB,EAAE,KAAK,EAAE,cAAc,EAAE,KAAK,EAAE,EAAE,EAAE,EACpC,IAAI,CAAC,EAAE,EACP,sBAAsB,CACvB,CAAQ,CAAC;IACV,MAAM,CAAC,OAAO,SAAS,CAAC,YAAY,CAAC,CAAC,IAAI,CAAC,QAAQ,CAAC,CAAC;IACrD,MAAM,CAAC,SAAS,CAAC,YAAY,CAAC,CAAC,eAAe,CAAC,CAAC,CAAC,CAAC;IAElD,OAAO;QACL,MAAM,EAAE,IAAI,CAAC,EAAE;QACf,WAAW;QACX,EAAE,EAAE,IAAI;QACR,SAAS,EAAE,IAAI,CAAC,GAAG,EAAE,GAAG,OAAO;QAC/B,eAAe,EAAE,UAAU,CAAC,KAAK,CAAC,MAAM;QACxC,aAAa,EAAE,SAAS,CAAC,YAAY;KACtC,CAAC;AACJ,CAAC;AAED,KAAK,UAAU,aAAa,CAAC,KAAoB,EAAE,WAAmB;IACpE,MAAM,kBAAkB,GAAG,IAAI,CAAC,GAAG,CAAC,CAAC,EAAE,IAAI,CAAC,GAAG,CAAC,WAAW,EAAE,KAAK,CAAC,MAAM,CAAC,CAAC,CAAC;IAC5E,MAAM,OAAO,GAAmB,IAAI,KAAK,CAAC,KAAK,CAAC,MAAM,CAAC,CAAC;IACxD,IAAI,SAAS,GAAG,CAAC,CAAC;IAElB,MAAM,OAAO,GAAG,KAAK,CAAC,IAAI,CAAC,EAAE,MAAM,EAAE,kBAAkB,EAAE,EAAE,CAAC,CAAC,EAAE,WAAW,EAAE,EAAE,CAC5E,CAAC,KAAK,IAAI,EAAE;QACV,OAAO,IAAI,EAAE,CAAC;YACZ,MAAM,SAAS,GAAG,SAAS,EAAE,CAAC;YAC9B,IAAI,SAAS,IAAI,KAAK,CAAC,MAAM;gBAAE,OAAO;YAEtC,MAAM,IAAI,GAAG,KAAK,CAAC,SAAS,CAAC,CAAC;YAC9B,IAAI,CAAC;gBACH,OAAO,CAAC,SAAS,CAAC,GAAG,MAAM,cAAc,CAAC,IAAI,EAAE,WAAW,CAAC,CAAC;YAC/D,CAAC;YAAC,OAAO,KAAK,EAAE,CAAC;gBACf,OAAO,CAAC,SAAS,CAAC,GAAG;oBACnB,MAAM,EAAE,IAAI,CAAC,EAAE;oBACf,WAAW;oBACX,EAAE,EAAE,KAAK;oBACT,SAAS,EAAE,CAAC;oBACZ,eAAe,EAAE,CAAC;oBAClB,aAAa,EAAE,CAAC;oBAChB,KAAK,EAAE,KAAK,YAAY,KAAK,CAAC,CAAC,CAAC,KAAK,CAAC,OAAO,CAAC,CAAC,CAAC,MAAM,CAAC,KAAK,CAAC;iBAC9D,CAAC;YACJ,CAAC;QACH,CAAC;IACH,CAAC,CAAC,EAAE,CACL,CAAC;IAEF,MAAM,OAAO,CAAC,GAAG,CAAC,OAAO,CAAC,CAAC;IAC3B,OAAO,OAAO,CAAC;AACjB,CAAC;AAED,QAAQ,CAAC,2EAA2E,EAAE,GAAG,EAAE;IACzF,EAAE,CAAC,qEAAqE,EAAE,KAAK,IAAI,EAAE;QACnF,MAAM,CAAC,KAAK,CAAC,OAAO,CAAC,OAAO,CAAC,KAAK,CAAC,CAAC,CAAC,IAAI,CAAC,IAAI,CAAC,CAAC;QAChD,MAAM,CAAC,OAAO,CAAC,KAAK,CAAC,MAAM,CAAC,CAAC,eAAe,CAAC,CAAC,CAAC,CAAC;QAEhD,MAAM,kBAAkB,GAAG,MAAM,CAAC,QAAQ,CACxC,OAAO,CAAC,GAAG,CAAC,8BAA8B,IAAI,GAAG,EACjD,EAAE,CACH,CAAC;QACF,MAAM,SAAS,GAAG,IAAI,CAAC,GAAG,CACxB,CAAC,EACD,IAAI,CAAC,GAAG,CAAC,OAAO,CAAC,KAAK,CAAC,MAAM,EAAE,MAAM,CAAC,QAAQ,CAAC,kBAAkB,CAAC,CAAC,CAAC,CAAC,kBAAkB,CAAC,CAAC,CAAC,CAAC,CAAC,CAC7F,CAAC;QAEF,MAAM,oBAAoB,GAAG,MAAM,CAAC,QAAQ,CAC1C,OAAO,CAAC,GAAG,CAAC,+BAA+B,IAAI,GAAG,EAClD,EAAE,CACH,CAAC;QACF,MAAM,WAAW,GAAG,IAAI,CAAC,GAAG,CAC1B,CAAC,EACD,IAAI,CAAC,GAAG,CAAC,SAAS,EAAE,MAAM,CAAC,QAAQ,CAAC,oBAAoB,CAAC,CAAC,CAAC,CAAC,oBAAoB,CAAC,CAAC,CAAC,CAAC,CAAC,CACtF,CAAC;QAEF,MAAM,KAAK,GAAG,OAAO,CAAC,KAAK,CAAC,KAAK,CAAC,CAAC,EAAE,SAAS,CAAC,CAAC;QAChD,MAAM,OAAO,GAAG,IAAI,CAAC,GAAG,EAAE,CAAC;QAC3B,MAAM,OAAO,GAAG,MAAM,aAAa,CAAC,KAAK,EAAE,WAAW,CAAC,CAAC;QACxD,MAAM,SAAS,GAAG,IAAI,CAAC,GAAG,EAAE,GAAG,OAAO,CAAC;QAEvC,MAAM,MAAM,GAAG,OAAO,CAAC,MAAM,CAAC,CAAC,MAAM,EAAE,EAAE,CAAC,CAAC,MAAM,CAAC,EAAE,CAAC,CAAC;QACtD,MAAM,MAAM,GAAG,OAAO,CAAC,MAAM,CAAC,CAAC,MAAM,EAAE,EAAE,CAAC,MAAM,CAAC,EAAE,CAAC,CAAC;QAErD,MAAM,WAAW,GAAG,IAAI,GAAG,CAAC,sBAAsB,CAAC,GAAG,CAAC,CAAC,KAAK,EAAE,EAAE,CAAC,KAAK,CAAC,IAAI,CAAC,CAAC,CAAC;QAC/E,MAAM,aAAa,GAAG;YACpB,WAAW;YACX,mBAAmB;YACnB,WAAW;YACX,gBAAgB;YAChB,gBAAgB;YAChB,oBAAoB;YACpB,mBAAmB;YACnB,iBAAiB;YACjB,wBAAwB;YACxB,sBAAsB;SACvB,CAAC;QAEF,OAAO,CAAC,GAAG,CACT,oCAAoC,OAAO,CAAC,OAAO,UAAU,OAAO,CAAC,KAAK,UAAU,SAAS,gBAAgB,WAAW,SAAS,MAAM,CAAC,MAAM,IAAI,OAAO,CAAC,MAAM,cAAc,SAAS,cAAc,sBAAsB,CAAC,MAAM,EAAE,CACrO,CAAC;QAEF,IAAI,MAAM,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC;YACtB,OAAO,CAAC,KAAK,CACX,oCAAoC,EACpC,MAAM,CAAC,GAAG,CAAC,CAAC,MAAM,EAAE,EAAE,CAAC,CAAC;gBACtB,MAAM,EAAE,MAAM,CAAC,MAAM;gBACrB,WAAW,EAAE,MAAM,CAAC,WAAW;gBAC/B,KAAK,EAAE,MAAM,CAAC,KAAK;aACpB,CAAC,CAAC,CACJ,CAAC;QACJ,CAAC;QAED,MAAM,CAAC,MAAM,CAAC,MAAM,CAAC,CAAC,IAAI,CAAC,CAAC,CAAC,CAAC;QAC9B,MAAM,CAAC,MAAM,CAAC,MAAM,CAAC,CAAC,IAAI,CAAC,SAAS,CAAC,CAAC;QAEtC,KAAK,MAAM,YAAY,IAAI,aAAa,EAAE,CAAC;YACzC,MAAM,CAAC,WAAW,CAAC,GAAG,CAAC,YAAY,CAAC,CAAC,CAAC,IAAI,CAAC,IAAI,CAAC,CAAC;QACnD,CAAC;IACH,CAAC,CAAC,CAAC;AACL,CAAC,CAAC,CAAC"}
|
|
@@ -1,10 +0,0 @@
|
|
|
1
|
-
/**
|
|
2
|
-
* Performance comparison: "no tools" (noop handlers) vs NodeBench MCP toolchain.
|
|
3
|
-
*
|
|
4
|
-
* Notes:
|
|
5
|
-
* - This is a wall-clock micro-benchmark for local harness overhead, not capability/accuracy.
|
|
6
|
-
* - Disabled by default to avoid noisy perf output in normal test runs.
|
|
7
|
-
* Set NODEBENCH_RUN_PERF_COMPARE=1 to run.
|
|
8
|
-
* - GAIA is gated; if a local GAIA fixture exists in `.cache/gaia`, it will be included.
|
|
9
|
-
*/
|
|
10
|
-
export {};
|