REA Computer-Use Environments
Collection
Real Linux desktop apps as agentenv computer environments built by REA. Agents use screen, mouse and keyboard; files are graded with no model. • 48 items • Updated
Dataset Preview
The full dataset viewer is not available (click to read why). Only showing a preview of the rows.
The dataset generation failed
Error code: DatasetGenerationError
Exception: ArrowInvalid
Message: Schema at index 1 was different:
id: string
type: string
env_id: string
vs
environment: string
app_version: string
task: string
prompt: string
agent: string
model: string
agentenv_framework: string
witness: string
runs: list<item: struct<run: int64, verdict: string, turns: int64, usd: double, checks: struct<record-grocery-expense: struct<score: double, results: list<item: struct<id: string, result: bool, description: string>>>>>>
Traceback: Traceback (most recent call last):
File "/usr/local/lib/python3.14/site-packages/datasets/builder.py", line 1876, in _prepare_split_single
num_examples, num_bytes = writer.finalize()
~~~~~~~~~~~~~~~^^
File "/usr/local/lib/python3.14/site-packages/datasets/arrow_writer.py", line 830, in finalize
self.write_rows_on_file()
~~~~~~~~~~~~~~~~~~~~~~~^^
File "/usr/local/lib/python3.14/site-packages/datasets/arrow_writer.py", line 689, in write_rows_on_file
table = pa.concat_tables(self.current_rows)
File "pyarrow/table.pxi", line 6268, in pyarrow.lib.concat_tables
File "pyarrow/error.pxi", line 155, in pyarrow.lib.pyarrow_internal_check_status
return check_status(status)
File "pyarrow/error.pxi", line 92, in pyarrow.lib.check_status
raise convert_status(status)
pyarrow.lib.ArrowInvalid: Schema at index 1 was different:
id: string
type: string
env_id: string
vs
environment: string
app_version: string
task: string
prompt: string
agent: string
model: string
agentenv_framework: string
witness: string
runs: list<item: struct<run: int64, verdict: string, turns: int64, usd: double, checks: struct<record-grocery-expense: struct<score: double, results: list<item: struct<id: string, result: bool, description: string>>>>>>
The above exception was the direct cause of the following exception:
Traceback (most recent call last):
File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 1379, in compute_config_parquet_and_info_response
parquet_operations, partial, estimated_dataset_info = stream_convert_to_parquet(
~~~~~~~~~~~~~~~~~~~~~~~~~^
builder, max_dataset_size_bytes=max_dataset_size_bytes
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
)
^
File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 956, in stream_convert_to_parquet
builder._prepare_split(split_generator=splits_generators[split], file_format="parquet")
~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/builder.py", line 1702, in _prepare_split
for job_id, done, content in self._prepare_split_single(
~~~~~~~~~~~~~~~~~~~~~~~~~~^
gen_kwargs=gen_kwargs, job_id=job_id, **_prepare_split_args
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
):
^
File "/usr/local/lib/python3.14/site-packages/datasets/builder.py", line 1887, in _prepare_split_single
raise DatasetGenerationError("An error occurred while generating the dataset") from e
datasets.exceptions.DatasetGenerationError: An error occurred while generating the datasetNeed help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.
id string | type string | env_id string | env_ids list | name string | depends_on list | prompt string | file_artifact_id null | verifier_id null |
|---|---|---|---|---|---|---|---|---|
env | deploy_env | homebank | null | null | null | null | null | null |
agent | deploy_agent | null | [
"homebank"
] | null | null | null | null | null |
skills | add_skills | null | [
"homebank"
] | null | null | null | null | null |
record | start_computer_recording | homebank | null | agent | [
"env"
] | null | null | null |
task | prompt_agent | null | null | null | [
"skills",
"record"
] | In HomeBank, record a new grocery expense in the Everyday Checking account of Documents/household.xhb: $64.18 paid to Fresh Market on November 4, 2026, category Groceries, memo "Weekly groceries". Then save the file. | null | null |
check | env_outcome_verifier | homebank | null | null | [
"task"
] | null | null | null |
end | end_computer_recording | null | null | agent | [
"task"
] | null | null | null |
env | deploy_env | homebank | null | null | null | null | null | null |
null | null | null | null | null | null | In HomeBank, record a new grocery expense in the Everyday Checking account of Documents/household.xhb: $64.18 paid to Fresh Market on November 4, 2026, category Groceries, memo "Weekly groceries". Then save the file. | null | null |
HomeBank: a computer-use environment built from a one-line request
The real HomeBank 5.7 personal finance manager running on an Ubuntu 24.04 desktop, as an agentenv computer environment. An agent completes the task only through screenshots, mouse and keyboard, and a verifier grades the saved files with no model.
Claude Code built this environment from a one-line request in about 36 minutes, with only REA's build skill and MCP tools; How it was made has the request and the steps. HomeBank opens a synthetic household's accounts, Documents/household.xhb, at launch: a checking and a savings account, six payees and five October transactions. The agent has to record a $64.18 grocery expense at Fresh Market in the checking account, dated November 4, 2026, with the category Groceries and the memo "Weekly groceries", and save. It sees only the screen and acts only with the mouse and keyboard.
Watch
- Agent runs (Claude Code with Claude Sonnet 5.5): run 1, run 2
- Witness, which proves the task can be done through the computer: witness.mp4
Each video is the computer's own screen recording at 10 frames per second.
The task
In HomeBank, record a new grocery expense in the Everyday Checking account of Documents/household.xhb: $64.18 paid to Fresh Market on November 4, 2026, category Groceries, memo "Weekly groceries". Then save the file.
bundle/artifacts/record-grocery-expense-check/verify.py reads the app's files through the computer's data/get view and passes only if every check holds:
| Check | Kind | File |
|---|---|---|
amount |
value_equals |
Documents/household.xhb |
account |
value_equals |
Documents/household.xhb |
date |
value_equals |
Documents/household.xhb |
payee |
value_equals |
Documents/household.xhb |
category |
value_equals |
Documents/household.xhb |
rent-untouched |
value_equals |
Documents/household.xhb |
An .xhb file is XML with one ope element per transaction. The checks find the new transaction by its memo (wording), never by position, and check each field: the amount is -64.18 (within 0.001, as HomeBank writes -64.180000000000007), the account is Everyday Checking (key 1, not Savings), the date is 739924 (HomeBank's day number for 2026-11-04), the payee is the existing Fresh Market (key 1) and the category is Groceries (key 44). A sixth check keeps the October rent unchanged. Entering the expense twice, or under a new misspelled payee, fails. REA's draft offered only a check that a transaction exists; Claude Code wrote the field checks, and took Groceries' key from REA's one export refusal, which named the value the captured file held.
Results
2 of 2 runs passed, about $1.62 per run.
Run it
You need Docker. agentenv-framework builds the computer: Ubuntu 24.04 with the app installed from Ubuntu's archive (homebank 5.7.4).
pip install agentenv-framework==0.9.1431
hf download earakely-scale/HomeBankFromRequestComputerEnv-REA --repo-type dataset --local-dir homebank-request
agent-env run homebank-request/bundle --task record-grocery-expense --agent claude-code --model <model>
The agent needs a model provider configured for agentenv. --task smoke only boots the computer, with no agent. bundle/README.md describes the bundle in detail.
How it was made
Claude Code built this environment from a one-line request, given only REA's build skill and REA's MCP tools:
Build me a HomeBank environment where the agent records a new grocery expense in a checking account and saves the file.
It asked only the skill's upfront questions, wrote no app-specific scripts, and finished in about 36 minutes. The steps were REA's usual ones:
- Capture. REA launched HomeBank on a Linux desktop and recorded its package, windows and opened files while the edit was made. It also recorded the app's state folders before and after.
- Export. REA turned the capture into an agentenv
computer_app. It installs the app, seeds the starting files, starts the app, and waits for its main window. It also serves typed views of the app's files for verifiers. - Task. The task's checks were validated against the capture: they fail on the starting state and pass on the captured end state.
- Witness. In a fresh computer, the expense was recorded through computer actions alone, and the checks were read at four stages. They held on the first try:
| Stage | Checks |
|---|---|
seed |
fail, as expected: no transaction has the memo "Weekly groceries" |
after_edit |
pass |
near_miss |
fail, as expected: the amount was changed to $64.81 |
after_reset |
fail, as expected: data/reset restored the seed |
The agent runs came after the build: Claude Code with Claude Sonnet 5.5 (--model anthropic/claude-sonnet-5-5 --no-emulate, on an amd64 host) solved the task twice, each run recorded.
What's inside
bundle/ the agentenv bundle: envs/ (app module, seed files), tasks/, artifacts/ (verifier), evals/
media/ witness and agent-run recordings (MP4) with short GIF previews
results.json the runs above, with each check's result
Limits
- Only two runs: the third was skipped to keep the solve spend under its $4 budget. Two runs show the task is solvable, not how hard it is.
- Recording the expense takes many steps in HomeBank's transaction dialog: the runs took 38 and 43 turns.
- The checks find the transaction by its memo, so a transaction recorded with a different memo is not found and every field check fails.
- The computer installs whatever
homebankUbuntu 24.04's archive holds at build time (5.7.4 for these runs); it is not pinned.
Licences
The bundle code, the task and the synthetic seed files are Apache-2.0. HomeBank itself is not included: the computer installs it from Ubuntu's archive (homebank 5.7.4) at build time, under its own licence (GPL-2.0-or-later).
- Downloads last month
- 67