Skip to content

feat(catalogue): add OSWorld2 with OSWorld-V2's own agents, its services on the machine and a null control - #134

Draft
sirotaoleg wants to merge 8 commits into
ThinkFlowLab:mainfrom
sirotaoleg:feat/osworld2
Draft

sirotaoleg wants to merge 8 commits into
ThinkFlowLab:mainfrom
sirotaoleg:feat/osworld2

Conversation

@sirotaoleg

@sirotaoleg sirotaoleg commented Nov 15, 2026 •

Copy link
Copy Markdown

Stacked on #133: this pull request's own commits are the last five. Until #133 is merged, the diff shows its files too. It can be merged after #133.

What

OSWorld-V2's 108 tasks become a bundled benchmark, osworld2, on the Nullius standard, at the upstream's release osworld-v2.1. Each task is a unit in an Ubuntu VM of its own on the upstream's docker provider, set up by the task's own class and scored from 0 to 1 by its own evaluate(). The tasks and their assets are gated on Hugging Face. Nullius does not fetch them, and the entry is gated (#133): a unit's failure from a model endpoint says its status alone, and the driver names an upstream exception by a known class's name at most, never by its words. The Original setup runs one of OSWorld-V2's own agents, M3 or Muse Spark, chosen by the Agent option, from its published script, with every model call sent to the relay. The Original's models are pinned to their providers on OpenRouter: M3 to MiniMax, Muse Spark to Meta. Claude Sonnet 4.6 is pinned to Anthropic; it is the LLM grader of 21 tasks and the simulated user of 6. Only M3's runs count as comparable with the paper's row. Its Review row and its generated page name the setup "OSWorld-V2 agents". A null control runs one DONE action, then the task's evaluate() on the untouched VM. The entry declares agent.computer/v1. No agent from the Agents page runs it yet. Its texts are one Good to know line and 25 limitations. Settings lists GitLab's root token under Custom keys, for the two GitLab tasks.

The mock websites of 40 tasks are OSWorld-web at the same tag. The guide's steps build them from source and serve them on the machine the units run on. A unit that needs them is refused before its VM starts when they do not answer. For 026 and 041, the driver starts GitLab anew from its snapshot, one task at a time, and Overleaf's 2 tasks get Overleaf started anew. Each of the 8 tasks that ask for a proxy gets a proxy of its own. The options are Steps (100, 300 or 500), Thinking for M3, the website host suffix, the GitLab host and its write share, and disk write limits for each VM. By default a run takes 100 steps with thinking off, as the upstream's M3 script does; the paper ran M3 at 500 steps with thinking on. Muse Spark runs as published at this release, while its published score came from newer upstream code; a limitation says what differs. There are six datasets: All tasks (108, the upstream's test_v2.json), Offline (82), and four by the services their tasks need. Tasks 103 and 104 are not supported: their evaluate() needs FreeCAD on the driver's machine. Review and the run's start ask for Docker and for the scoring programs (ImageMagick, ffmpeg, tesseract) on the driver's machine. As for OSWorld1, the Where step greys out the container card under the Original, whose docker provider needs Docker on the machine. A unit that ends rate-limited is not run again by itself (limits.run_again: false): an episode is long and paid. A guide, docs/benchmarks/osworld2-machine.md, prepares a machine, runs the null control on one task as its last setup step, and says what a run takes in time and money.

Touches

catalogue · templates and static (the mark) · docs and site (docs/benchmarks/, the generated pages, AGENTS.md) · routes (tests only) · packaging (the entry's .yml, .Dockerfile and startup/*.py as package data in pyproject.toml)

Before and after

Before After
Benchmarks page before
The Benchmarks page before this change
Benchmarks page after
OSWorld2 listed, with its mark
Benchmark step before
The builder's Benchmark step under Computer use: OSWorld1, beside the selected GAIA
Benchmark step after
OSWorld1 and OSWorld2
Agents page before
The Agents page: "n/7 benchmarks it runs"
Agents page after
"n/8": OSWorld2 counts, and no agent runs it
Custom keys before
Settings, Custom keys: "No custom keys yet."
Custom keys after
OSWorld2's GITLAB_PRIVATE_TOKEN, not set
Codex's page before
An agent's page, Codex: the benchmarks it runs
Codex's page after
OSWorld2 listed: it needs agent.computer/v1, which Codex does not offer
New OSWorld2's page
OSWorld2's page: its setup and its proofs by dataset
New OSWorld2's options
Its options: the agent, steps, thinking, the VM provider and its services
New OSWorld2's dataset
Its dataset and its Good to know line
New OSWorld2's limitations
The first of its limitations
New Dataset step
Dataset step: six datasets with their controls, and GitLab's token not set
New Agent step
Agent step: OSWorld-V2 agents; the Agents page's agents lack computer use; the settings every agent shares
New Agent step, the Original's settings
The Original's own settings: M3 or Muse Spark, steps, thinking
New Agent step with Muse Spark
Muse Spark chosen: Thinking, an M3 setting, is gone
New Model step
Model step, Like the paper: MiniMax-M3 on OpenRouter, Claude Sonnet 4.6 as the judge
New Where step
Where step: no container backend here; Docker and the scoring programs checked on this machine
New Review step
Review: every decision, both prerequisites ready, no model key in this fresh profile
New Null control job
The null control's merged job on test_v2, its first rows
New Job of 052
052 by the test stand's model in M3: 1.0

Taken at 1400 px, light theme, on a fresh profile. Cut to part of their page: the Benchmark step, Agents and Settings pairs (the top 1000 px), Codex's page (the top 1100 px), OSWorld2's page (four parts) and the null control's job (its first rows). The two jobs are runs named under Verified, opened in the app. The Settings pair, the Model step and Review were taken again after main's Settings icons and Hosts (#163). After the rebase onto main at 5bcfd82d, Settings and the builder's steps were taken on main and on this branch and compared with them: only the Model step changed, where #124 as merged shows Served by as one row, and it was taken again at 592be82f.

Verified

  • At 592be82f, on feat(standard): let a benchmark keep its tasks gated, pin its own providers, set its own limits and prove a merged null control #133 (a6d32e74) on main at 5bcfd82d, on Python 3.12: backend suite (python3 app/tests/parallel.py), 2561 tests in 141 modules, OK; route suite with Node on PATH, 491 tests in 44 modules, OK; the wheel's check (app/packaging/verify.py), OK.
  • OSWorld2's 15 test modules, 371 tests, OK; they need the entry, so none passes without it. 10 are skipped here: they run only when the environment names the upstream with its gated task classes, OSWorld-web or GitLab. Three route test modules and three backend test modules changed for the new entry: the benchmark lists and counts, the Where step's container card, Settings' custom keys, and OSWorld2's cases of gated tasks and pinned routes (test_standard_gated.py, its tests in the class Definition of test_standard_relay_pin.py).
  • In a headless Chromium, light and dark, 1400 and 390 px: the Benchmarks page, OSWorld2's page, the Agents page, Settings and the builder, no page error or cut text. /ui-sweep --only osworld2 at fc0bdb1f, this pull request's third commit before the rebase (573e8d09 now, the entry's files unchanged): 6 screenshots, none with a page error or cut text.
  • PYTHONPATH=app python3 -m nullius standard check: ok. python3 app/tests/catalogue_docs.py --write, then the check: up to date. python3 app/tests/field_contract.py: ok. app/site/readme_media.py --check: current. The docs check reports the same two findings in AGENTS.md as on main.
  • Where the runs below ran. The commits named here are of the port's working branches, not in this repository:
    • c3637ffd is this pull request's tree with main's two later commits merged in;
    • f74d1a00 and 58c2b586 add one test commit: two JSON files, no code. It gives OSWorld2's Original one more model, MiniMax M3.1 Flash, the name it is served as, its pin to MiniMax and --max_tokens 32768. That model is a test stand, not in this pull request;
    • 58c2b586 is this entry on the core before feat(standard): choose the serving providers that may answer a model, in the builder and on the command line #124;
    • 34e35370 and 8e2e381b are of the port before its move to this repository's history.
  • The controls, recorded in standard.json's proven:
    • Offline, no services, on 2026-11-04, on an earlier commit: the 36 ids are the same twice, and the null control scores 0 on 4 tasks;
    • All tasks, on 2026-11-15, on fc0bdb1f, this pull request's third commit before the rebase (573e8d09 now); the fourth only writes this record, and the fifth changes three tests. The null control ran on all 108 tasks, one launch with each task a shard of its own, at 16 MB/s and 160 IOPS a VM, in six lanes with at most seven VMs at a time; it took 1 h 32 min. 108 and 093 failed their setup under that load and passed alone at 24 MB/s and 240 IOPS, in 24 more minutes. The grader cost $0.14. nullius standard prove osworld2 --dataset test_v2 --job nop=<merged job> passed: the 108 ids are the same twice, 0 of 108 at full score, a mean of 0.0021 against a bound of at most 0.01. Five tasks score above 0 on the untouched VM, as a limitation says;
    • every task is in line with the same control on all 108 tasks on 34e35370, which had the same mean; 077's LLM grader gave 0.054 both times.
  • An agent's score on every kind of task, each above the same task's null control (0.0 in each):
    • Seven kinds, on 58c2b586, by the test stand's model in OSWorld-V2's M3 agent, from the command line:

      • desktop apps, 045: 1.0;
      • mock websites, 052: 1.0;
      • Overleaf, 072: 0.775;
      • GitLab, 026: 0.5, a partial score;
      • a proxy, 075: 1.0 (recorded 0.9999999999999999), with the unit's proxy carrying new outbound connections while the agent acted;
      • the public internet, 027: 1.0;
      • the LLM grader, 003: 1.0, with both grader calls answered 200.

      Each of the seven gives its recorded score again at c3637ffd when its outcome is computed from its saved records, with no VM and no evaluate().

    • The eighth, the simulated user: 007, 0.8, by Muse Spark on 8e2e381b; the simulated user answered 7 times. It was not run on this base.

  • Live on this base, against OpenRouter:
    • on c3637ffd, the null control on 087, a graded task: 0.0. The grader's call went to Anthropic by the definition's pin, and standard.served reads held: true;
    • on f74d1a00, the test stand's model on 052: 1.0. All 68 calls went to MiniMax by the test commit's pin and named it, held: true, with no error.
  • This pull request's own agents have no positive score on this base. M3 ran one task here, 017, on the core before feat(standard): choose the serving providers that may answer a model, in the builder and on the command line #124 (90fde73e), and scored 0.0: the agent declared the task failed at its 120th step. Muse Spark ran none. Their scores before the move, on 34e35370 and earlier:
    • M3 on 9 tasks: 070 1.0 twice, 030 1.0, 045 1.0, 013 1.0 then 0.0, 031 0.714, 017 0.429, 010 0.0 three times and once an environment error; 026 and 098 left no score (a full context window, a spend guard);
    • Muse Spark on 3 tasks: 045 1.0, 007 0.8; 052 left no score (a launch error of ours, then a spend guard).
  • Every attempt before the move, two test stands' models included, was recorded and triaged: 61 attempts on 33 tasks. 43 were scored (21 full, 11 partial, 11 zero), 10 ended in an environment error and 8 were stopped by a guard or by hand. None was an agent error. Each of their full and partial scores was checked for the false-success patterns the port's audits found: none.
  • Not run: a positive score of M3 or Muse Spark on this base; the simulated user on this base; task 041 with an agent; a parity run of the paper's 108 tasks; the 10 skipped tests with the upstream named; the docker tier; the site build.

Risk

  • OSWorld2's driver and OSWorld-V2's scripts run only when someone runs OSWorld2. The Agents page counts eight agent benchmarks, and the generated agent pages gain an OSWorld2 column. No agent there declares agent.computer/v1. Settings shows OSWorld2's GitLab token under Custom keys to every user.
  • Scores. A run compares with the paper's row by its model alone: Docker instead of AWS, no US residential proxy, release osworld-v2.1, the grader through OpenRouter, and by default 100 steps with thinking off. A wrong error kind would move a unit between "environment error" (left out of the adjusted rate) and the agent's error (counted).
  • The machine.
    • VMs, the mock websites, GitLab and the proxies run on the machine the units run on, outside a run's constraints in Settings; -j bounds them.
    • GitLab's Docker in Docker and the websites' image build run as privileged containers.
    • In each of the driver's processes, startup/sitecustomize.py wraps the creation of the VM's container: its network, its ports on 127.0.0.1, its disk and its write limits. The VMs need a free Docker subnet.
    • The VM's clock is UTC+8: a dated task set up before 16:00 UTC and acted on after can score 0.
  • Secrets. GitLab's root token is a licence of the benchmark, handed to the driver only. Model keys stay with the relay. Three tasks' mail profiles lose their saved sign-ins before the VM starts.
  • Stacked on feat(standard): let a benchmark keep its tasks gated, pin its own providers, set its own limits and prove a merged null control #133. Once it is merged, this branch is rebased onto main.
  • Undone by reverting the commits; the catalogue entry goes with them.

harenome added a commit that referenced this pull request Nov 16, 2026
Closes #136

## What

On Runs → By launch, an agent or control with 2 or more shards is now
one row, when `cli merge` would merge those shards. The row shows the
numbers of the merged job. `launches.fold` makes that job in memory: the
trials of every shard, and the job record from `merge.result_of`.
`result_of` is the code that `merge` used inline to compute the job
record, now a shared function. `merge.check` keeps the trial records
that it already reads.

- The rate is over every trial with a verdict, never an average of shard
rates.
- Units that no shard ran come from `merge.check` and stand beside the
count ("2 of 8 not run"). They are never misses. The notes of `merge`
are the tooltip.
- The state is the most telling state of the shards, with a warning such
as "1 of 2 shards stopped".
- The header gives the dataset size, not the size of one shard.

The row is folded by default. A click on the row, or Enter or Space on
its chevron, unfolds one row per shard ("shard 2 · host-2"), each with
its Open. The fold uses the shared chevron and slide of
`disclosures.js`, and respects reduced motion. A refresh of the list
keeps an unfolded row unfolded. Controls fold the same way, stay
unranked and stay out of Compare. The task-check line is judged over
every shard: "Task checks incomplete: oracle not run on shard 2."

A participant with one shard keeps its plain row. Shards that `merge`
refuses also keep their rows. Job pages, `cli merge` and merged jobs do
not change. The header counts every shard ("8 runs"), every shard has
its row, and each folded row says how many shards it holds ("2 shards").

A fold reads every trial record of its shards. It is kept in memory for
the shard jobs that the workspace cache gives, and it is read again only
when the workspace parses a shard again. A new set of shards replaces
each kept fold that shares a shard with it, so the cache holds one fold
per shard at most.

## Touches

backend · the standard (`merge.py`) · routes · templates and static ·
docs and site

## Before and after

One launch over 2 machines, 8 units: 2 agents, nop and oracle. Shard 2
stopped while terminus-2 ran (2 of its 4 units), and the oracle did not
run on shard 2. Light theme, 1280 px.

| Before | After, folded |
| --- | --- |
| <img width="980" height="670" alt="1-before-shards-apart"
src="https://github.com/user-attachments/assets/990ceda8-8212-4003-ab1c-f380be783fc9"
/> | <img width="980" height="421" alt="2-after-folded"
src="https://github.com/user-attachments/assets/f37d9af1-7b3b-4b05-abc6-7dca2da1c983"
/> |

Before: 8 rows, 2 rates for each agent, "4 tasks", "oracle not run".
After: 4 rows, "8 tasks". terminus-2 shows 50% and "3/6 · 2 of 8 not
run" with "1 of 2 shards stopped". The task-check line reads "Task
checks incomplete: oracle not run on shard 2."

| After, unfolded (click and keyboard) | After, dark, 390 px |
| --- | --- |
| <img width="980" height="595" alt="3-after-unfolded"
src="https://github.com/user-attachments/assets/b30f3189-d737-4658-8701-a3bbc434ad4a"
/> | <img width="358" height="844" alt="4-after-dark-390"
src="https://github.com/user-attachments/assets/6c4b106c-218b-4308-a934-ecac1f96934d"
/> |

## Verified

- Backend test `Fold.test_a_fold_reads_as_the_merged_job`: three shards,
one absent, and one with a planned unit that did not run. The folded job
equals `parse_job` of the `merge.merge` output on every field the row
shows (solved, rates, outcomes, planned, errors, start, end, time,
tokens, cost, mean score, budget). The rate is not the mean of the shard
rates. The units not run and the notes are those of `merge`. Shards that
`merge` refuses are not folded.
- Backend test
`Fold.test_a_fold_is_read_again_only_when_the_workspace_parses_a_shard_again`:
a second fold of the same cached jobs does not call `merge.check`. After
a trial record changes, it calls `merge.check` again. After a third
shard appears, the cache holds only the fold of the three.
- Route tests `test_launch_shards`: 8 shards give "8 runs", 8 shard rows
and 4 folded rows that each say "2 shards". The rows are folded by
default. Each shard row has its Open, and the folded row has none. The
rate and time of the folded row equal those of the merged job. "2 of 8
not run" and "1 of 2 shards stopped" show. The controls fold and stay
out of Compare. A nop that passes one task on shard 2 gives "A control
did not behave". A stopped nop shard gives "nop stopped on shard 2".
Participants with one shard keep their plain rows.
- On `c7af06d5`: backend suite `python3 app/tests/parallel.py` (1989
tests) and route suite `python3 app/tests/parallel.py app/ui_tests` (328
tests) pass in a clean container.
- Browser, with the fixture above: folded and unfolded, light and dark,
1280 and 390 px. No page errors and no horizontal page scroll. A click
on the row unfolds it, and the shard rows slide open and closed. Tab
moves the focus to the chevron of the oracle, Enter unfolds it, and
Space folds it again. With reduced motion, the rows show at once. At 390
px, a long agent name wraps and the chevron stays on the line of the
name. The name, its "2 shards" line and the shard rows start at the same
left edge.
- Render time of `/runs?view=launch`, median of 15 requests after a
first one, with the same parsed jobs given to the page. 2 agents and nop
and oracle, each 4 shards of 500 units (8000 trials): 71 ms before this
change, 101 ms after. The first render reads the trial records once: 540
ms, against 199 ms before. The rest of the added time is the outcome
bars of the 4 folded rows, over 2000 trials each. 8 shards of 1 unit
each (32 trials): 11 ms before, 13 ms after.

## Risk

If the fold is wrong, a row shows numbers that differ from the merged
job. The backend test compares the two directly. `merge` writes the same
`result.json` as before through `result_of`, and its tests pass
unchanged. The first render of a large sharded launch reads all its
trial records once (+0.3 s for 8000 trials). Later renders use the cache
until a shard changes. The cache holds one fold per shard at most, with
the parsed jobs that the workspace cache also holds. Open pull requests
#124, #133 and #134 also change `merge.py`: the second to merge must
rebase. Revert the commit to undo.
harenome added a commit that referenced this pull request Nov 16, 2026
…l everywhere (#148)

Closes #147

## What

Each model provider is now one record that every feature reads. A
feature asks for the protocol that it speaks, and gets the providers
that serve it.

- **The registry** (`app/nullius/providers.py`). `PROTOCOLS` defines
each wire format once: its default path and how the key is sent (`chat`,
`responses`, `messages`, `systemone`, `decisions`, `openai-decisions`).
`PRESETS` holds OpenRouter, OpenAI, Anthropic and TypeSafe. `record()`
returns `{name, base, key, endpoints}` for a preset or a custom
provider. `key` is the name of the secret setting, never the key.
- **Storage** (`app/nullius/settings.py`). A record is one JSON setting
of a new kind, `provider`. A custom provider is `MODEL_<ID>`. The
endpoints added to a preset are `OPENROUTER`, `OPENAI`, `ANTHROPIC` or
`TYPESAFE`. For these settings the environment does not replace the
file. Its value is a partial object, merged over the file's value (RFC
7396, `null` removes a field). Settings shows read-only only the fields
that the environment sets. Keys keep their names and stay separate
secrets.
- **Migration.** `MODEL_<ID>_NAME` and `MODEL_<ID>_API_URL` are still
read as `name` and `base`. Saving a card moves them into the record.
`nullius migrate --providers` previews that rewrite of `settings.json`,
and `--apply` writes it. Recipes do not change. `local` and `local:<ID>`
still read as `custom`.
- **Consumers.** The registry is read by the builder, the provider test,
onboarding, `execution`, standard runs, the relay, the CLI's
`--provider` and both judges. The builder's JavaScript copy of the list
is gone. The builder offers a provider only where it serves what the
recipe's agents and judge speak, so Anthropic appears for an agent of
the standard that speaks `messages`. Review, Run (`web/execute.py`) and
`run.Run` refuse a provider that does not fit. The relay sends
`x-api-key` and `anthropic-version` on a `messages` route only, and a
bearer header on a benchmark's own routes.
- **Settings cards.** A card shows Base and a collapsed Endpoints
section with one row for each protocol. The placeholder is the default
URL. A base on a preset's host suggests that preset's endpoints. A
preset's known endpoints are fixed. An endpoint added to a preset is a
path, or a full URL on the preset's own host, because the preset's key
goes there.
- **System One** (`app/nullius/judge/client.py`) asks each decision
protocol: the TypeSafe shape on `systemone` and `decisions`, and its own
adapter for `openai-decisions`. It keeps the provider's `usage.cost` and
records a `refusal` as no answer. A 402 says what it means. A judge
never follows a redirect with the key.
- **Judge cards.** Each judge has Provider, then Protocol when the
provider serves two protocols of that judge (else one quiet line, "via
..."), then Model. Each judge's provider is a select whatever the count,
because the list is the user's own providers and grows or shrinks with
them; cards are kept for fixed sets of choices. New settings:
`LLM_JUDGE_PROTOCOL`, `SYSTEM_ONE_PROTOCOL`. On OpenRouter, System One
offers OpenRouter's public list of decision models, asked without a key,
cached 6 hours, with the fixed list as the fallback.

The seven commits are atomic and in reading order: registry and storage,
consumers, Settings cards, decision protocols, judge cards, tests, docs.

## Touches

backend (`providers`, `settings`, `migrate`, `onboarding`, `execution`,
`judge/client`, `recipes/launch`) · the standard (`run`, `relay`,
`recipe`, `cli`, `definitions`, `fidelity`) · routes (`web/settings`,
`web/execute`, `web/filters`) · templates and static (Settings cards,
judge cards, builder core) · docs (`docs/settings/model-providers.md`,
`system-one.md`, `llm-judge.md`, `README.md`, `agent-keys.md`,
`docs/builder.md`, `docs/judgement.md`, `docs/standard.md`,
`docs/cli/README.md`, `AGENTS.md`, `app/.env.example`)

## Before and after

Settings at 1280 px, light theme. Fixture keys only.

Model providers. TypeSafe is a preset like the others, so it comes after
the run providers.

| Before | After |
| --- | --- |
| <img width="748" height="431" alt="1-before-providers"
src="https://github.com/user-attachments/assets/1dd2338e-0b32-4631-8895-a6df2ba7fc29"
/> | <img width="748" height="431" alt="2-after-providers"
src="https://github.com/user-attachments/assets/ff3bb210-22ef-4bff-bbe4-df4ac4488519"
/> |

A preset card (OpenRouter), Endpoints open. Known endpoints are fixed.
Other protocols are editable rows.

| Before | After |
| --- | --- |
| <img width="748" height="360" alt="3-before-preset-card"
src="https://github.com/user-attachments/assets/5fb776e9-1c86-4cf0-8508-24041b6f9042"
/> | <img width="748" height="706" alt="4-after-preset-card"
src="https://github.com/user-attachments/assets/9058e8e4-339e-49da-8a44-67b2ffcdf8bf"
/> |

A custom provider, Endpoints open, with one endpoint set.

| Before | After |
| --- | --- |
| <img width="748" height="451" alt="5-before-custom-card"
src="https://github.com/user-attachments/assets/c725f642-3d22-4b8d-8fb0-c702e415594d"
/> | <img width="748" height="897" alt="6-after-custom-card"
src="https://github.com/user-attachments/assets/afcbcb01-fed9-4ef3-87c2-883d73594edc"
/> |

Judges. System One on OpenRouter serves two of its protocols, so it
shows Protocol. The LLM judge on OpenAI serves one, so it shows "via
OpenAI Chat Completions".

| Before | After |
| --- | --- |
| <img width="748" height="504" alt="7-before-judges"
src="https://github.com/user-attachments/assets/a089a1bd-9f72-4ca6-bff1-e1a652a83ec1"
/> | <img width="748" height="503" alt="8-after-judges"
src="https://github.com/user-attachments/assets/61ef9ae3-e080-4e54-99a1-dc43d4960ba6"
/> |

The custom card at 390 px:

<img width="358" height="1153" alt="9-after-custom-card-390"
src="https://github.com/user-attachments/assets/dce35d02-44b1-4ccb-8362-c84129e1eb87"
/>

## Verified

The branch is based on `main` at 1cc37ac. I tested its tip, each suite
in a new rootless Podman container (`python:3.12`, `--init`, uv on PATH,
Node for the route suite):

- Backend suite (`python3 app/tests/parallel.py`): 2029 tests in 118
modules, OK. `main` at 1cc37ac, in the same container: 1992 tests in
116 modules, OK.
- Route suite (`python3 app/tests/parallel.py app/ui_tests`): 441 tests
in 42 modules, OK. `main` at 1cc37ac: 431 tests in 41 modules, OK.
- New modules: `test_provider_registry` (17 tests: the merge, `null`,
partial environment values, the old settings, record checks, the preset
host rule, keys never in a record), `test_judge_decisions` (18 tests:
each decision protocol's request and reply, `usage.cost`, `refusal`,
402, no redirect, the cached model list and its fallback),
`ui_tests.test_judge_cards` (2 tests: the protocol row and the live
list). Existing modules gained tests for the relay's headers, the
builder's provider filter, Harbor recipes on Anthropic and the Settings
cards.
- End to end, against the server (`app/app.py`) and the CLI, in a
container with no network, with a new `NULLIUS_HOME` for each case and
fixture keys only. 37 checks, all pass:
  - No settings: the four presets and the OpenAI-compatible card.
- A custom provider in `settings.json`: its card, its merged record, the
default path for a protocol without an entry, System One offers it for
its `decisions` endpoint, no key in the page.
- `MODEL_GW='{"base": "https://gw-eu.example/v1", "endpoints":
{"decisions": null}}'` in the environment: the base changes, the name
stays the file's, the endpoint goes, System One no longer offers it. The
card shows Base read-only and Name editable. The file does not change.
- `nullius migrate --providers`: the preview names the moves, prints no
value and leaves the file as it was. `--apply` writes the record,
removes the old lines, keeps the key and the file mode 0600. A second
run finds nothing to move.
- Old variables in the environment (`MODEL_API_URL`, `MODEL_<ID>_NAME`,
`MODEL_<ID>_API_URL`, `LLM_JUDGE_PROVIDER=local:<ID>`): read as before.
- System One's model list on OpenRouter, offline: the route answers with
the built-in list in less than the 5 s timeout, or with a cached list,
however old.
- Browser (Chromium, Playwright), on this branch and on `main`, at 1280
px and 390 px, light and dark, with no network: the provider list, a
preset card and a custom card with Endpoints open, and both judge cards.
No page errors and no horizontal scroll.

Not verified: live calls to OpenRouter Decisions and OpenAI Decisions,
and a whole run on Anthropic. The tests use fixtures.

## Risk

A wrong record or header can send a call to the wrong URL, or with the
wrong key header. Undo with a revert. The old settings are still read,
so a revert keeps working with them. Records written by a card stay in
`settings.json` and the older code ignores them.

- **Migration on edit.** Saving a custom provider's card moves
`MODEL_<ID>_NAME` and `MODEL_<ID>_API_URL` into `MODEL_<ID>`. After a
revert, that provider loses its name and base until they are set again.
`nullius migrate --providers` without `--apply` shows what would move.
- **New reserved names.** `OPENROUTER`, `OPENAI`, `ANTHROPIC`,
`TYPESAFE` and `MODEL_<ID>` are now settings. An agent key listed under
one of these names is left out, and Settings says so. A shell variable
such as `MODEL_PATH` that is not a JSON object is not read as a record.
- **`--provider anthropic` on the CLI.** The key header now follows the
agent's protocol. An agent that speaks only `chat` is refused on
Anthropic, where it used to get `x-api-key` on OpenAI-style paths.
- **Runs and custom endpoint paths.** The relay forwards the agent's own
path below the provider's base. An Endpoints row changes where the
judges and the test ask, not where a run's calls go.
- **`openai-decisions` field names** come from OpenAI's documentation,
not from a live reply.
- **Open PRs on the same files.** This change does not depend on any of
them. Whichever merges second rebases. A test merge of this branch with
each one:
- #124 (OpenRouter serving providers, `--served-by`): conflicts in
`providers.py`, `standard/run.py`, `standard/relay.py` and
`docs/settings/model-providers.md`. Each conflict is small. Both add
functions at the same place in `providers.py`. This branch removes
`run.PROVIDERS`, where #124 adds `routing` to the OpenRouter entry: that
flag becomes a field of the OpenRouter preset. Both change
`Relay._headers`.
- #133 (stacked on #124): the same four files, and also
`standard/fidelity.py` (one `key_header=` line, now `r.key_header`) and
`docs/standard.md` (one table row each).
- #134 (stacked on #133): the same as #133, and also
`ui_tests/test_settings_page.py` (one placeholder assertion, which both
change).
- #124, #133 and #134 also conflict with `main` alone in
`standard/merge.py` (and #133, #134 in `tests/test_standard_shard.py`),
which this branch does not touch.
…on top of --served-by: gated tasks, a definition's pins as its serving pins, a pinned route's refusal as the unit's, limits of their own, and prove of a merged control

A definition's pin (setups.original.pin, score.judge.pin) joins the run's --served-by pins and wins over
--served-by for its model; ApiProviderRouteError reads the pin and reply.served_by. A failed Responses
object reads reply.end and reply.error.
…eep the score of a unit a guard stopped despite a last 429

limits.run_again false: a unit that ended rate-limited is not run once more; its outcome stands. A unit whose driver scored it after a guard's stop keeps that score, marked stopped, though its last own call got a 429.
… own last 429, not a side call's or a context window's stop

The rule reads the agent's own last call, a guard's line in the stop file and a score stopped by other than the context window; a cap's refusal is still the cap's. limits.run_again takes null as unset. The rate limits row names the opt-out.
… upstream's own runners, mock websites, GitLab and proxy, proven on test_v2, and how to prepare a machine for it

OSWorld2 runs each of the 108 tasks of test_v2 in a VM of its own through the upstream's docker provider, with the mock websites its tasks need served on the machine, and GitLab and a proxy started for each unit that needs one. Its Original runs the upstream's agents with their models pinned to their providers on OpenRouter, and its null control is proven on test_v2 at 0 of 108. The tasks are gated, so a failing endpoint is kept as its status alone. Its machine page says what a machine needs, how to check it and what a run costs.
limits.run_again false: an OSWorld2 episode is paid and long; one that ended rate-limited is run again by hand.
…g programs, and fake the sweep's clock for the sweep alone

Two tests made a Run of OSWorld2, which asks for ImageMagick, ffmpeg and
tesseract on the machine: they now say the programs are there, as the
test of their refusal does. The sweep's tests replaced time.sleep for
every module, so a slow docker's wait in subprocess advanced the fake
clock and the test's own steps: the fake now replaces the sweep's time
module alone.

@aitianshi aitianshi left a comment

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

GitLab for tasks 026 and 041 should be supported automatically, or left out of this PR.

Right now it is half done. Each unit starts GitLab from its snapshot and removes it afterwards, which is automatic. The one-time preparation is not: docs/benchmarks/osworld2-machine.md step 9 asks the user to clone Task-Web/gitlab at the pinned commit, run gitlab_server.py configure, compose pull and init, choose a root token, paste it into Custom keys as GITLAB_PRIVATE_TOKEN, and set GitLab host. A user should not have to set up a GitLab server by hand to run a dataset.

The builder makes this visible on every card. It shows Add GITLAB_PRIVATE_TOKEN in Custom keys (some tasks may need it). even on "Offline, no services", whose 36 tasks never touch GitLab. That happens because no dataset names an environment, so credentialWarnings treats the gitlab environment's secret as possibly needed everywhere.

Please choose one of these:

  1. Support it automatically. Nullius prepares GitLab on the run machine as part of the run's prerequisites (nullius/prerequisites.py): clone at the pin, configure, pull, init. It generates the root token there and keeps it on that machine, so the user never types or pastes one. The builder then shows nothing about GitLab unless the preparation fails.
  2. Remove it from this PR. Drop the GitLab parts (gitlab_server.py, gitlab.compose.yml, the gitlab environment and option, the licence, step 9 of the doc, its tests). Tasks 026 and 041 then stay out of the offered datasets until a follow-up PR supports them fully.

Either way, a selection that cannot need the token should not warn about it.

"good_to_know": [
"Scores are partial credit: each task's own evaluate() gives 0 to 1, and the upstream's score is the mean."
],
"limitations": [

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Limitations are far too long. Please cut them down a lot.

There are 25 items and about 2,200 words; one item alone (the LLM grader) is 332 words. On OSWorld2's page they fill several screens, and nobody will read them before choosing a run.

The UI rules (.claude/rules/ui.md) say Good to know and limitations must change a user's choice, the action they have to take, or how they read the results. They also say to leave out dependency and installation details that Nullius handles itself, and that removing text beats adding it. Most of these items are how the driver works inside, how errors are classified, or how the machine is set up:

  • installation and machine setup: mock websites build, GitLab, proxy image, VM resources, write limits, Settings constraints, output kept per trial. These belong in docs/benchmarks/osworld2-machine.md, or nowhere if Nullius does them;
  • outcome mechanics: how a setup failure, a timeout, the upstream's own failures, the 10-hour stop, a grader refusal or a context overflow is scored. These belong in the docs or in the trial's own record.

What a user actually needs is roughly this, one short sentence each:

  • not the paper's setup: another release, no residential proxy, and 100 steps with thinking off by default (Steps 500 and Thinking adaptive restore the budget);
  • Muse Spark's published score came from newer upstream code and does not compare;
  • the grader and the simulated user are Claude Sonnet 4.6 on OpenRouter: with another provider those tasks fail;
  • tasks that use the public internet can change outcome with the live sites;
  • tasks 103 and 104 are not supported;
  • a slower machine can lower scores;
  • the null control scores above 0 on five tasks.

Aim for under 10 items, each one line.

@aitianshi

Copy link
Copy Markdown
Collaborator

The screenshots in the description can't be read without opening them, and opening them fails.

Each image is shown at width="420", so the text in it (the limitations, the options) is too small to read and the reader has to click it. Clicking it opens the private attachment link. For a reviewer here that gave an empty file: "The file 707192284-87ab46cf-...png could not be opened because it is empty" (the "OSWorld2's limitations" screenshot).

Please put the screenshots directly in the description at full size, for example by dragging them into the editor and leaving GitHub's own ![...](...) line without width. Crop each one to the part that matters, so it can be read on the page without clicking a link.

"websites": {
"label": "Website host suffix",
"summary": "WEBSITE_HOST_SUFFIX: none, or the host suffix OSWorld-web's sites are served under for this machine's VMs, a wildcard DNS name of the address the stack listens on, the gateway of the VMs' own network (10.213.0.1.nip.io). The 40 tasks that use the sites need it.",
"pattern": "none|(?=.{3,200}$)[a-z0-9](?:[a-z0-9-]{0,61}[a-z0-9])?(?:\\.[a-z0-9](?:[a-z0-9-]{0,61}[a-z0-9])?)+",

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The page should show an example for these options, not the regex.

OSWorld2's page prints the raw pattern of all five free-text options (Website host suffix, GitLab host, GitLab write share, Disk write MB/s, Disk write IOPS). It shows none|(?=.{3,200}$)[a-z0-9](?:[a-z0-9-]{0,61}... where a user needs to see what to type. catalogue/benchmark.html only hides the pattern for temperature ("Custom decimal, e.g. 0.5"); every other pattern option falls through to the raw regex. Before this PR no other option had a pattern, so nobody saw it.

Suggested:

  • each pattern option carries an example, e.g. "example": "10.213.0.1.nip.io", "gitlab.10.213.0.1.nip.io", "16:160", "16", "160";
  • the page shows "e.g. 16:160" instead of the pattern, the same way temperature already does, so the special case goes away;
  • the pattern stays for validation only;
  • the summaries drop the internal names they start with (WEBSITE_HOST_SUFFIX:, GITLAB_URL:, device_write_bps), following the "Names, not keys" rule in .claude/rules/ui.md.

},
"websites": {
"label": "Website host suffix",
"summary": "WEBSITE_HOST_SUFFIX: none, or the host suffix OSWorld-web's sites are served under for this machine's VMs, a wildcard DNS name of the address the stack listens on, the gateway of the VMs' own network (10.213.0.1.nip.io). The 40 tasks that use the sites need it.",

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Please make the option and setup descriptions much shorter.

On the page, each option's description runs 2 to 4 lines (35 to 44 words for Website host suffix, GitLab host, GitLab write share and Disk write MB/s). The two Setup blocks are longer still (Docker is 54 words). They explain how things work inside: Caddy, snapshots, device_write_bps, 1024 * 1024 bytes, the network nullius-osworld2. A user only needs to know what the value does and when to set it. The UI rules (.claude/rules/ui.md): "Less text", "Removing text beats adding it", and the details go on the entry's own docs page.

For example:

  • Website host suffix: "Where the mock websites are served. Needed by the 40 website tasks."
  • GitLab host: "Where GitLab is served. Needed by tasks 026 and 041."
  • GitLab write share: "Disk write limit for GitLab, as MB/s:IOPS."
  • Disk write MB/s / IOPS: "Disk write limit for each VM."
  • Setup: Docker: "Docker without sudo; KVM recommended."
  • Setup: scoring programs: "ImageMagick, ffmpeg and tesseract, used to score 6 tasks."

Two related things on the same page:

  • the Setup headings show raw keys (Setup: docker, Setup: scoring_programs) because benchmark.html prints the key. They need a label, per "Names, not keys";
  • the thinking option has no label.

This branch has not been deployed

No deployments
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants