Awesome

WorkArena: A Benchmark for Evaluating Agents on Knowledge Work Tasks

[Benchmark Contents] ♦ [Getting Started] ♦ [Live Demo] ♦ [BrowserGym] ♦ [Citing This Work]

Papers

[ICML 2024] WorkArena: How Capable are Web Agents at Solving Common Knowledge Work Tasks? [Paper]
[NeurIPS 2024] WorkArena++: Towards Compositional Planning and Reasoning-based Common Knowledge Work Tasks [Paper]

WorkArena is a suite of browser-based tasks tailored to gauge web agents' effectiveness in supporting routine tasks for knowledge workers. By harnessing the ubiquitous ServiceNow platform, this benchmark will be instrumental in assessing the widespread state of such automations in modern knowledge work environments.

The preferred way to evaluate on WorkArena is with AgentLab which will conduct parallel experiments through BrowserGym and report on a unified leaderboard.

https://github.com/ServiceNow/WorkArena/assets/2374980/68640f09-7d6f-4eb1-b556-c294a6afef70

Getting Started

To setup WorkArena, you will need to get your own ServiceNow instance, install our Python package, and upload some data to your instance. Follow the steps below to achieve this.

a) Create a ServiceNow Developer Instance

Go to https://developer.servicenow.com/ and create an account.
Click on Request an instance and select the Washington release (initializing the instance will take a few minutes)
Once the instance is ready, you should see your instance URL and credentials. If not, click Return to the Developer Portal, then navigate to Manage instance password and click Reset instance password.
You should now see your URL and credentials. Based on this information, set the following environment variables:
- SNOW_INSTANCE_URL: The URL of your ServiceNow developer instance
- SNOW_INSTANCE_UNAME: The username, should be "admin"
- SNOW_INSTANCE_PWD: The password, make sure you place the value in quotes "" and be mindful of escaping special shell characters. Running echo $SNOW_INSTANCE_PWD should print the correct password.
Log into your instance via a browser using the admin credentials. Close any popup that appears on the main screen (e.g., agreeing to analytics).

Warning: Feel free to look around the platform, but please make sure you revert any changes (e.g., changes to list views, pinning some menus, etc.) as these changes will be persistent and affect the benchmarking process.

b) Install WorkArena and Initialize your Instance

Run the following command to install WorkArena in the BrowswerGym environment:

pip install browsergym

Then, install Playwright:

playwright install

Finally, run this command in a terminal to upload the benchmark data to your ServiceNow instance:

workarena-install

Your installation is now complete! 🎉

Benchmark Contents

At the moment, WorkArena-L1 includes 19,912 unique instances drawn from 33 tasks that cover the main components of the ServiceNow user interface, otherwise referred to as "atomic" tasks. WorkArena++ contains 682 tasks, each one sampling among thousands of potential configurations. WorkArena++ uses the atomic components presented in WorkArena, and composes them into real-world use cases evaluating planning, reasoning, and memorizing abilities of agents.

The following videos show an agent built on GPT-4-vision interacting with every atomic component of the benchmark. As emphasized by our results, this benchmark is not solved and thus, the performance of the agent is not always on point.

Knowledge Bases

Goal: The agent must search for specific information in the company knowledge base.

The agent interacts with the user via BrowserGym's conversational interface.

https://github.com/ServiceNow/WorkArena/assets/1726818/352341ba-b501-46ac-bfa6-a6c9be1ac2b7

Forms

Goal: The agent must fill a complex form with specific values for each field.

https://github.com/ServiceNow/WorkArena/assets/1726818/e2c2b5cb-3386-4f3c-b073-c8c619e0e81b

Service Catalogs

Goal: The agent must order items with specific configurations from the company's service catalog.

https://github.com/ServiceNow/WorkArena/assets/1726818/ac64db3b-9abf-4b5f-84a7-e2d9c9cee863

Lists

Goal: The agent must filter a list according to some specifications.

In this example, the agent struggles to manipulate the UI and fails to create the filter.

https://github.com/ServiceNow/WorkArena/assets/1726818/7538b3ef-d39b-4978-b9ea-8b9e106df28e

Menus

Goal: The agent must navigate to a specific application using the main menu.

https://github.com/ServiceNow/WorkArena/assets/1726818/ca26dfaf-2358-4418-855f-80e482435e6e

Dashboards

Goal: The agent must answer a question that requires reading charts and (optionally) performing simple reasoning over them.

Note: For demonstration purposes, a human is controlling the cursor since this is a pure retrieval task

https://github.com/ServiceNow/WorkArena/assets/1726818/0023232c-081f-4be4-99bd-f60c766e6c3f

Getting Started

To setup WorkArena, you will need to get your own ServiceNow instance, install our Python package, and upload some data to your instance. Follow the steps below to achieve this.

a) Create a ServiceNow Developer Instance

Go to https://developer.servicenow.com/ and create an account.
Click on Request an instance and select the Washington release (initializing the instance will take a few minutes)
Once the instance is ready, you should see your instance URL and credentials. If not, click Return to the Developer Portal, then navigate to Manage instance password and click Reset instance password.
You should now see your URL and credentials. Based on this information, set the following environment variables:
- SNOW_INSTANCE_URL: The URL of your ServiceNow developer instance
- SNOW_INSTANCE_UNAME: The username, should be "admin"
- SNOW_INSTANCE_PWD: The password, make sure you place the value in single quotes '' and be mindful of escaping special shell characters. Running echo $SNOW_INSTANCE_PWD should print the correct password.
Log into your instance via a browser using the admin credentials. Close any popup that appears on the main screen (e.g., agreeing to analytics).

b) Install WorkArena and Initialize your Instance

Run the following command to install WorkArena in the BrowswerGym environment:

pip install browsergym-workarena

Then, install Playwright:

playwright install

Finally, run this command in a terminal to upload the benchmark data to your ServiceNow instance:

workarena-install

Your installation is now complete! 🎉

Live Demo

Run this code to see WorkArena in action.

Note: the following example executes WorkArena's oracle (cheat) function to solve each task. To evaluate an agent, calls to env.step() must be used instead.

To run a demo of WorkArena-L1 (ICML 2024) tasks using BrowserGym, use the following script:

import random

from browsergym.core.env import BrowserEnv
from browsergym.workarena import ALL_WORKARENA_TASKS
from time import sleep


random.shuffle(ALL_WORKARENA_TASKS)
for task in ALL_WORKARENA_TASKS:
    print("Task:", task)

    # Instantiate a new environment
    env = BrowserEnv(task_entrypoint=task,
                    headless=False)
    env.reset()

    # Cheat functions use Playwright to automatically solve the task
    env.chat.add_message(role="assistant", msg="On it. Please wait...")
    cheat_messages = []
    env.task.cheat(env.page, cheat_messages)

    # Send cheat messages to chat
    for cheat_msg in cheat_messages:
        env.chat.add_message(role=cheat_msg["role"], msg=cheat_msg["message"])

    # Post solution to chat
    env.chat.add_message(role="assistant", msg="I'm done!")

    # Validate the solution
    reward, stop, message, info = env.task.validate(env.page, cheat_messages)
    if reward == 1:
        env.chat.add_message(role="user", msg="Yes, that works. Thanks!")
    else:
        env.chat.add_message(role="user", msg=f"No, that doesn't work. {info.get('message', '')}")

    sleep(3)
    env.close()

To run a demo of WorkArena-L2 (WorkArena++) tasks using BrowserGym, use the following script. Change the filter on line 6 to l3 to sample L3 tasks.

import random

from browsergym.core.env import BrowserEnv
from browsergym.workarena import get_all_tasks_agents
 
AGENT_L2_SAMPLED_SET = get_all_tasks_agents(filter="l2")
 
AGENT_L2_SAMPLED_TASKS, AGENT_L2_SEEDS = [sampled_set[0] for sampled_set in AGENT_L2_SAMPLED_SET], [
    sampled_set[1] for sampled_set in AGENT_L2_SAMPLED_SET
]
from time import sleep

for (task, seed) in zip(AGENT_L2_SAMPLED_TASKS, AGENT_L2_SEEDS):
    print("Task:", task)

    # Instantiate a new environment
    env = BrowserEnv(task_entrypoint=task,
                    headless=False)
    env.reset()

    # Cheat functions use Playwright to automatically solve the task
    env.chat.add_message(role="assistant", msg="On it. Please wait...")
    
    for i in range(len(env.task)):
        sleep(1)
        env.task.cheat(page=env.page, chat_messages=env.chat.messages, subtask_idx=i)
        sleep(1)
        reward, done, message, info = env.task.validate(page=env.page, chat_messages=env.chat.messages)
   
    if reward == 1:
        env.chat.add_message(role="user", msg="Yes, that works. Thanks!")
    else:
        env.chat.add_message(role="user", msg=f"No, that doesn't work. {info.get('message', '')}")

    sleep(3)
    env.close()

Note: the following example executes WorkArena's oracle (cheat) function to solve each task. To evaluate an agent, calls to env.step() must be used instead.

Citing This Work

Please use the following BibTeX to cite our work:

WorkArena

@misc{workarena2024,
      title={WorkArena: How Capable Are Web Agents at Solving Common Knowledge Work Tasks?}, 
      author={Alexandre Drouin and Maxime Gasse and Massimo Caccia and Issam H. Laradji and Manuel Del Verme and Tom Marty and Léo Boisvert and Megh Thakkar and Quentin Cappart and David Vazquez and Nicolas Chapados and Alexandre Lacoste},
      year={2024},
      eprint={2403.07718},
      archivePrefix={arXiv},
      primaryClass={cs.LG}
}

WorkArena++

@misc{boisvert2024workarenacompositionalplanningreasoningbased,
      title={WorkArena++: Towards Compositional Planning and Reasoning-based Common Knowledge Work Tasks}, 
      author={Léo Boisvert and Megh Thakkar and Maxime Gasse and Massimo Caccia and Thibault Le Sellier De Chezelles and Quentin Cappart and Nicolas Chapados and Alexandre Lacoste and Alexandre Drouin},
      year={2024},
      eprint={2407.05291},
      archivePrefix={arXiv},
      primaryClass={cs.AI},
      url={https://arxiv.org/abs/2407.05291}, 
}