Internet measurement faces twin challenges: complex analyses require expert-level orchestration of tools, yet even syntactically correct implementations can have methodological flaws and can be difficult to verify. Democratizing measurement capabilities thus demands automating both workflow generation and verification against methodological standards established through decades of research. We present Airavat, the first agentic framework for Internet measurement workflow generation with systematic verification and validation. Airavat coordinates a set of agents mirroring expert reasoning: three agents handle problem decomposition, solution design, and code implementation, with assistance from a registry of existing tools. Two specialized engines ensure methodological correctness: a Verification Engine evaluates workflows against a knowledge graph encoding five decades of measurement research, while a Validation Engine identifies appropriate validation techniques grounded in established methodologies. Through four Internet measurement case studies, we demonstrate that Airavat (i) generates workflows matching expert-level solutions, (ii) makes sound architectural decisions, (iii) addresses novel problems without ground truth, and (iv) identifies methodological flaws missed by standard execution-based testing.
Figures & tables
Figure 1. Airavat’s architecture comprising a multi-agent workflow generation pipeline, verification engine, validation engine, knowledge graph, and registry.
Case Study
Problem
LoC
Key Capability
Results
1: Cable Impact
SeaMeWe-5 failure
∼ 850
Expert Solution Replication
Matches the expert solution
2: Disaster
Earthquake/hurricane
∼ 700
Workflow Simplicity
Appropriate expert-level minimal solution
3: Cascading
Europe-Asia cables
∼ 1600
9-function orchestration
Multi-layer integration framework
4: Prefix2Org
Prefix-to-org mapping
1500 → 1700
Domain Transferability +
0% → 70-75% (vague query);
(w/ verification)
Verification capabilities
0% → 90.9% tags (improved query)
Table 1. Case Study Summary
Figure 2. Country-level impact from SeaMeWe-5 cable failure. The heatmap shows the risk factor, computed as the fraction of IP addresses affected in a country relative to the total number of IP addresses mapped to that country. The results match the findings of Xaminer ( Ramanathan et al., 2024 ) .
Figure 3. Global cable infrastructure impact from earthquakes and hurricanes. The heatmap shows the risk factor, computed as the fraction of IP links affected in a country relative to the total number of IP links in that country. The results match the findings of Xaminer ( Ramanathan et al., 2024 ) .
Existing browser agent benchmarks face a fundamental trilemma: real-website benchmarks lack reproducibility due to content drift, controlled environments sacrifice realism by omitting real-web noise, and both require costly manual curation that limits scalability. We present WebForge, the first fully automated framework that resolves this trilemma through a four-agent pipeline -- Plan, Generate, Refine, and Validate -- that produces interactive, self-contained web environments end-to-end without human annotation. A seven-dimensional difficulty control framework structures task design along navigation depth, visual complexity, reasoning difficulty, and more, enabling systematic capability profiling beyond single aggregate scores. Using WebForge, we construct WebForge-Bench, a benchmark of 934 tasks spanning 7 domains and 3 difficulty levels. Multi-model experiments show that difficulty stratification effectively differentiates model capabilities, while cross-domain analysis exposes capability biases invisible to aggregate metrics. Together, these results confirm that multi-dimensional evaluation reveals distinct capability profiles that a single aggregate score cannot capture. Code and benchmark are publicly available at https://github.com/yuandaxia2001/WebForge.
Reliable evaluation of AI agents operating in complex, real-world environments requires methodologies that are robust, transparent, and contextually aligned with the tasks agents are intended to perform. This study identifies persistent shortcomings in existing AI agent evaluation practices that are particularly acute in web agent evaluation, as exemplified by our audit of WebVoyager, including task-framing ambiguity and operational variability that hinder meaningful and reproducible performance comparisons. To address these challenges, we introduce Emergence WebVoyager, an enhanced version of the WebVoyager benchmark that standardizes evaluation methodology through clear guidelines for task instantiation, failure handling, annotation, and reporting. Emergence WebVoyager achieves an inter-annotator agreement of 95.9%, indicating improved clarity and reliability in both task formulation and evaluation. Applying this framework to evaluate OpenAI Operator reveals substantial performance variation across domains and task types, with an overall success rate of 68.6%, substantially lower than the 87% previously reported by OpenAI, demonstrating the utility of our approach for more rigorous and comparable web agent evaluation.
Reusable Web-agent verification requires a rule to follow changing requirements and support subsequent execution. We introduce AgentProofBench, which connects paired tests of a frozen acceptance rule to execution from identical saved openings. The resource spans 97 tasks, 79 families, and four Web services. Experiments with DeepSeek Flash/Pro establish two linked measurements. Across 48 families, complete-panel and decision-replay tests retain 82/96 Flash and 79/96 Pro source programs; paired reuse-rebuild BA intervals are [-5.47, 1.57] and [-10.16, 0.00] percentage points. Across 22 execution families, deployment-state panel correctness separates program assignments whose two saved openings are both delivered correctly: 32/33 assignments with error-free panels, versus 1/11 with panel errors. All reference-defined program-gate regressions and withheld deliveries occur in the latter group. Neutral continuation attains the highest correct delivery under the tested generic feedback, providing a concrete baseline for measuring the value added by a gate. Released instructions, states, programs, reference rules, and trajectories connect each test failure to its execution consequences. AgentProofBench makes reusable verification assessable at both the requirement and policy levels.