WorkArena

workarena · v1.0.0

web gui

WorkArena evaluates agents on enterprise service-desk workflows inside a real ServiceNow Personal Developer Instance. Tasks are organized into three levels: L1 atomic tasks (~33 unique tasks x multiple seeds), L2 compositional tasks, and L3 extended tasks with company protocols. Requires a free ServiceNow PDI and browser-automation tooling.

By: @younik (Omar Younis) , @NicolasAG (Nicolas Gontier)

Install

pip install workarena-cube

Version: 1.0.0 · PyPI page

333

Tasks

local

Infra

Yes

Debug Task

Yes

Debug Agent

Feature Flags

— async

— streaming

— multi_agent

— multi_dim_reward

Legal

Wrapper license MIT

Benchmark license

Apache-2.0 Self-reported — verify before commercial use Source →

Notices

Software Registration

Requires a free ServiceNow Personal Developer Instance (PDI). Register at developer.servicenow.com

More info →

License information is self-reported by the cube developer and has not been verified by the AI Alliance. Always consult the source URL and original benchmark authors for authoritative terms.

Slow check not yet run. Stress test results will appear here after the async compliance check completes.

Parallelization

Mode

task-parallel

Registry Entry (YAML)

View on GitHub →

id: workarena
name: "WorkArena"
version: "1.0.0"
description: >
  WorkArena evaluates agents on enterprise service-desk workflows inside a
  real ServiceNow Personal Developer Instance. Tasks are organized into three
  levels: L1 atomic tasks (~33 unique tasks x multiple seeds), L2
  compositional tasks, and L3 extended tasks with company protocols. Requires
  a free ServiceNow PDI and browser-automation tooling.
package: workarena-cube
dev_install_url: "git+https://github.com/The-AI-Alliance/cube-harness#subdirectory=cubes/workarena"

authors:
- github: younik
  name: Omar Younis
- github: NicolasAG
  name: Nicolas Gontier

legal:
  wrapper_license: MIT
  benchmark_license:
    reported: Apache-2.0
    source_url: "https://github.com/ServiceNow/WorkArena/blob/main/LICENSE"
    verified_by_original_authors: false
  notices:
  - type: software_registration
    description: "Requires a free ServiceNow Personal Developer Instance (PDI). Register at developer.servicenow.com"
    url: "https://developer.servicenow.com"

tags:
- web
- gui

paper: "https://arxiv.org/abs/2402.05181"
parallelization_mode: task-parallel

status: active
task_count: 333
has_debug_task: true
has_debug_agent: true
resources: []
action_space: []
features:
  async: false
  streaming: false
  multi_agent: false
  multi_dim_reward: false