SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2410.06703 · ICLR · 2024

ST-WebAgentBench: A Benchmark for Evaluating Safety and Trustworthiness in Web Agents

Ido Levy, Ben Wiesel, Sami Marreed, Alon Oved, Avi Yaeli, Nir Mashkif, Segev Shlomov

arXiv · PDF · Open in the Atlas

Code that ran

1 of 1 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 1 failure of our instrument; 0 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2410.06703.json · badge: /api/ran/2410.06703.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
segev-shlomov/ST-WebAgentBench community
Papers with Code archive
1 ran · 0 no recorded run
FunctionStatusWhere it lives
compute_completion_rates Ran
ran; our contract draft was wrong, not the code
segev-shlomov/ST-WebAgentBench/stwebagentbench/result_analysis/analyze.py
pointer only (licence: NOASSERTION) · get_code("a42579504bbb0797")
How it ran

The body of this sample is withheld: GitHub could not classify the licence of this paper's repository, segev-shlomov/ST-WebAgentBench (NOASSERTION). The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

Also present in other papers' code

Of 1 samples lifted from this paper's linked repositories, 1 carry a function identity and 0 do not. Of the 1 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 1 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Autonomous web agents solve complex browsing tasks, yet existing benchmarks measure only whether an agent finishes a task, ignoring whether it does so safely or in a way enterprises can trust. To integrate these agents into critical workflows, safety and trustworthiness (ST) are prerequisite conditions for adoption. We introduce ST-WEBAGENTBENCH, a configurable and extensible framework designed as a first step toward enterprise-grade evaluation. Each of its 375 tasks carries one or more ST policies (3,057 in total), concise rules encoding constraints, and is scored along six orthogonal dimensions (e.g., user consent, robustness). Tasks span three difficulty tiers for fine-grained capability profiling, and a "Modality Challenge" disentangles vision-only from DOM-only information retrieval, isolating the contribution of each perceptual modality to agent failures. Beyond raw task success, we propose the Completion Under Policy (CuP) metric, which credits only completions that respect all applicable policies, and the Risk Ratio, which quantifies ST breaches across dimensions. Evaluating three open state-of-the-art agents shows their average CuP is less than two-thirds of their nominal completion rate, revealing substantial safety gaps. To support growth and adaptation to new domains, ST-WebAgentBench provides modular code and extensible templates that enable new workflows to be incorporated with minimal effort, offering a practical foundation for advancing trustworthy web agents at scale.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2410.06703")
get_code_for_paper("2410.06703")
have("2410.06703")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2410.06703.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2410.06703.svg)](https://syntology.ai/paper/2410.06703)

Connect an agent — have() is free.