Commit Graph
1200 Commits
Author SHA1 Message Date
Devika Verma 2d08fa2f0b fix(models): a ticked row's name on the provider page opens the model's settings (OPE-252)
The unverified note said "set the window under the model's settings" but the
page had no way there: rows were not clickable, and an unticked model is not
in the composer's picker either. A ticked row's name is now a button that opens
the Pick a model dialog on that model, like the picker card does; the note
says "tick the model, then click its name to set it" (en, zh).
2026-11-17 18:38:30 +05:30
Devika Verma e4fc1a80b5 fix(models): live rows read like the reviewed ones; the attachment tag says 'reads images and PDFs' (OPE-252)
Seen on the Gemini page: reviewed rows read "Gemini 3.6 Flash · Google" while
the rows from Google's own list read "Gemma 4 31B IT". A live row now carries
the suffix the reviewed rows of that provider use ("Google", "Anthropic",
"via OpenRouter"; the descriptor's title for a provider with no reviewed
row), on the page and in the composer's labels.

"images + PDF" read like a list, not a capability: "reads images and PDFs",
"reads images", nothing for text only, with the hover line saying so.
2026-11-17 18:38:30 +05:30
Devika Verma dcdf80978b fix(models): one list on the provider page; image, music, video and robotics models are not offered as coworker models (OPE-252)
Owner's reading of the Gemini page: no split between reviewed rows and the
rest (one list, reviewed first, the provider's newest after, new ones above
under their own heading), and a question about Nano Banana Pro, Lyria and
Robotics-ER showing up with a tick box.

Google serves its image, music, video and robotics models through the same
generateContent method as its language models, so the method alone did not
tell them apart. The chat-only filter now reads ids and display names for
those families too. They are not coworker models: a coworker's turns need a
model that holds a conversation and calls tools, and a picker row for a
generator would only fail in a session. Generation as a capability belongs
to a tool, not to the model picker.

A listed id that is not a chat model is still "on the key": a user who types
it by hand is not told the key lacks it (the listing keeps every id for that
check and shows only the chat ones).
2026-11-17 18:38:30 +05:30
Devika Verma 3defd442df fix(models): the provider page says what it means: Reviewed by OpenWorker / Also on your key, '200K context', 'images + PDF' (OPE-252)
Owner's reading of the first build: 'Recommended' hid that the split is reviewed versus not reviewed, a bare '200K' looked like a guess, and 'images' said nothing about PDF. Rows now carry the catalog's pdf flag; the tags read '200K context' and 'images + PDF' (or 'images'), each with a hover line saying where the figure comes from.
2026-11-17 18:38:30 +05:30
Devika Verma 68c2e5258c fix(models): a dated id folds into its catalog row; cloud windows print as the vendor writes them (OPE-252)
Seen on the first run against a real Anthropic key: the list carries
claude-haiku-4-5-20251001 while the catalog and the API know claude-haiku-4-5,
so three curated rows read "not on this key" and their dated twins appeared
again under All models. A dated id now folds into the undated row: one row,
on the key, with the listing's facts and date; add_model accepts either
spelling. A dated id with no catalog twin stays its own row.

The page printed 200,000 as "195K" (the local servers' power-of-two steps).
Cloud windows are decimal: 200K, 922K, 1M.
2026-11-17 18:38:30 +05:30
Devika Verma e23c974770 feat(models): a keyed provider's page lists what its own model list says the key can call (OPE-252)
Layer 1 of OPE-252. Until now a provider's page offered only the curated rows,
and a new model had to be typed in by id with nothing checking it.

Backend
- providers/live_models.py asks the configured provider's own model list with
  the user's key (Anthropic and Gemini their way, Azure at its endpoint, the
  OpenRouter sign-in at OpenRouter, every OpenAI-compatible vendor at its base
  URL) through the shared probe client, reads the three answer shapes in use,
  keeps chat models only, newest first, and carries whatever facts the list
  gives: context window, tool and image support, release date. The ChatGPT
  plan, Bedrock, Vertex and Ark have no list and say so.
- GET /v1/providers/{name}/catalog: the listing merged with the curated rows,
  cached a day per provider (Refresh asks again). Each row says where its facts
  come from (catalog, provider, unverified), whether it is in the picker,
  recommended, new since the check before, and on the key at all. "New" is
  what the list gained since the previous check at least an hour old, so two
  refreshes in a row do not swallow the badge; the first look marks nothing.
  A failed re-ask keeps the last list and says why; a failed first ask falls
  back to the curated rows.
- add_model refuses an id the key's own list does not carry ("Not available on
  this key") and names the closest ids; nothing changes for local servers,
  providers without a list, or before a list was asked.
- The listing's facts for picker models the catalog does not know go to
  live_facts.json; the engine's window resolution and capabilities_for read
  them. Settings answers model_verification per picker model and extends
  labels and windows with the listing's. A window the user saved now beats
  the catalog's (it used to lose).
- The overflow learner: when a provider refuses a long prompt and names its
  limit, the engine saves that figure as the model's context window when ours
  was unknown or larger (compaction.overflow_limit reads the figure from the
  OpenAI, Anthropic, vLLM and generic wordings).

GUI
- ProviderModels on a keyed provider's page: "N models on this key · checked
  2 min ago · Refresh", then New since last check, Recommended, All models
  (searchable, folds at 12), each row a tick for the picker with its context,
  images, new / unverified / not on this key tags and Make default. A typed id
  the key cannot call shows the refusal and "Did you mean" buttons. Providers
  without a list keep the curated checklist under the reason.
- The provider card shows "N new" when the list gained models.
- The model settings dialog lets the user set the context window of a cloud
  model the catalog does not know, and says whether the figure came from the
  provider's list or nowhere.

Tests: tests/test_live_models.py (45: the answer shapes incl. Gemini, Together,
Mistral, Cohere, Groq and garbage; every keyed provider has a request or a
reason; the environment key; fetch never raises; the cache, Refresh, the
restart, "new" across the hour rule, failures; add_model refusal with
suggestions and the cases it must not gate; the facts reaching settings,
the engine and capabilities_for; the window order; the overflow learner;
the route), overflow_limit in test_compaction, ProviderModels.test.tsx (8)
and ProviderGroups.badge.test.tsx. Backend suites around it: 230 passed;
whole GUI suite 769 passed; tsc clean.
2026-11-17 18:38:30 +05:30
Devika Verma 8df2d747f2 test(models): the move to model_catalog.json changed no fact (OPE-252)
A snapshot of MATRIX as the hand-written matrix.py served it on 2026-11-07, 104 rows with every id, label, capability and window; test_model_catalog.py checks the loaded matrix against it while the file still carries the migration serial. The first catalog change bumps the serial and retires the gate.
2026-11-17 18:38:30 +05:30
Devika Verma 9d472d1840 feat(models): the model matrix is a data file (OPE-252)
The 95 hand-written rows of `MATRIX` move from matrix.py to
coworker/providers/model_catalog.json: per provider and model, the label,
tools / vision / pdf / parallel_tool_calls, the context window, and an optional
output limit. Each provider block carries a `note` naming the vendor page or
listing its rows were read from and when (the comments matrix.py used to hold).
A row a maintainer sized on purpose (a window pinned to a particular host, a
capability unverified over this backend) says so with `pinned` and a reason in
its note; 13 rows are pinned.

matrix.py loads the file into the same `MATRIX` dict at import; `entry_for`,
`model_labels`, `model_context_windows` and `models_for_provider` are unchanged,
so no consumer changes. The signed-in OpenRouter provider stays a load-time
mirror and is never listed. `OVERRIDES` (empty) beats the file, before and after
the mirror. `reload()` refills the dict in place under a lock, the helpers read
a snapshot, and a file that fails validation leaves MATRIX untouched.

The loader refuses anything it cannot read as a whole: wrong schema, unknown or
missing fields, a non-boolean capability, a context that is not a positive
whole number, an OpenAI id with a prefix, a mirrored provider listed by hand,
a pin on a field that cannot be pinned. The file is in canonical form
(json.dumps indent 2, one trailing newline) so a catalog pull request diffs
only the rows that changed.

Keeping the file current is a maintainer job, not a runtime one: the app never
fetches a catalog from anyone. `serial`, `generated` and `last_checked` are
informational to the app and move when the maintainers change or check a fact.

Round trip against the hand-written dict: the same 104 ids (95 rows plus the
9 openrouter-account mirrors) in the same order, with the same labels,
capabilities and windows.

Tests: tests/test_model_catalog.py (file contract, package data, the four
helpers against the file, the mirror row for row, capabilities_for, reload in
place, a refused file leaves MATRIX, OVERRIDES on both twins, four reader
threads over 120 reloads, 34 refused shapes, the optional fields accepted);
test_matrix_current_models reuses matrix.full_id. The 251 existing
matrix-reading tests pass unchanged.
2026-11-17 18:38:30 +05:30
Devika Verma 96ef2d07f6 fix(ollama): a profile saved as the old localhost default becomes 127.0.0.1, once (OPE-250)
The Settings form prefills the Ollama address with the default and saves
every field, so anyone who saved the Ollama card before OPE-229 (2026-11-02)
carries `http://localhost:11434` without having typed it. On Windows
`localhost` tries ::1 first, where Ollama does not listen, so every new
connection (each probe, each chat request) waited 0.3-2 s. With the rest of
this branch in place that profile alone kept a cold Models & Keys load at
1.0 s; with it migrated the same load takes 0.39 s.

At manager start, a saved `base_url` that is exactly the old default (a
trailing slash or `/v1` allowed, case-insensitive host) is rewritten to
`DEFAULT_OLLAMA_URL`, keeping the `/v1` and the rest of the profile. OPE-229's
rule that a saved address is used as typed stands for anything else:
`localhost:11435`, a path after the port, or another host are left alone.

Test: the three migrated spellings, three addresses that must stay, and the
no-profile case.

(cherry picked from commit 5a94252ee4f528a8a8e1f200730e558320b3790e)
2026-11-17 18:38:05 +05:30
Devika Verma 53b4a8eb30 perf(models): one HTTP client for every local-server probe; a server that does not answer is not asked for its models (OPE-250)
Measured on Windows 11 with Ollama (5 models) and no other local server
running: a cold GET /v1/providers took 55 s on main, and 51 s again once
the one-minute probe caches lapsed; the composer's cold /v1/settings took
1.9 s.

Three costs stacked on top of the ones the branch already removed:

- httpx.get() builds a new client per call, and building one loads the
  system certificate store: about 2 s on Windows, for a plain http request
  to this machine that never uses TLS. Every probe (local_server.py,
  ollama_facts.py, the manager's Ollama liveness check) now goes through
  one client, providers/probe_http.py, built once at manager start, off the
  request thread.
- A server that did not answer was asked a second time, for its models.
  get_providers and the parallel warm-up list models only for servers that
  are up.
- Four local-server defaults still said `localhost`, which Windows resolves
  to ::1 first; they name 127.0.0.1 now, as Ollama's does since OPE-229.
  Ollama's per-model /api/show calls get the same short loopback connect
  as the listing: with a profile saved as `localhost` each one waited about
  2 s on the ::1 attempt.

Add button: reads "Adding…" and is disabled until the save returns; a
failed save shows the server's reason under the row and keeps the typed
id.

After: cold /v1/providers 1.0 s (0.35 s with the 127.0.0.1 default), 0.8 s
once the caches lapse, cold /v1/settings 0.26 s.

Tests: test_providers_page_speed.py covers the shared client, the
no-second-probe rule and the defaults; ModelChecklist.test.tsx the Add
feedback. Fixtures that stubbed httpx.get now stand the shared client in.

(cherry picked from commit 50415bc08367b87f42b1328b64d155162319f661)
2026-11-17 18:38:04 +05:30
Rohit Prasad 21f0035151 Merge pull request #906 from andrewyng/feat/inspectionbuddy-example
Add InspectionBuddy sample app
2026-11-16 15:36:48 -08:00
Rohit P 51e79f4269 fix: keep InspectionBuddy chat independently scrollable 2026-11-16 15:10:46 -08:00
Rohit P 8f7138349a feat: add InspectionBuddy edge sample with local setup 2026-11-16 14:58:54 -08:00
Rohit Prasad f91a610069 Merge pull request #905 from andrewyng/release/sdk-packaging-20261116
Fix SDK release packaging
2026-11-16 09:58:13 -08:00
Rohit P 469c4e769f Build Windows release runtimes with a static CRT 2026-11-16 05:50:02 -08:00
Rohit P d0bd8291fe Fix SDK release packaging and qualify builder installs 2026-11-16 05:09:20 -08:00
Rohit Prasad ecb445490f Merge pull request #898 from andrewyng/docs/desktop-getting-started
Add Desktop App docs
2026-11-13 16:34:18 -08:00
Rohit Prasad f68159ee2d Merge pull request #896 from andrewyng/docs/sdk-launch-guides
Update SDK docs and quickstart examples
2026-11-13 16:33:52 -08:00
Rohit P ad61ed95a9 Keep the cluster recording instructions in the guide 2026-11-13 16:07:35 -08:00
Rohit Prasad c5a04fa428 Merge pull request #895 from andrewyng/fix/cluster-installation-release
Add cluster UI controls and deployment examples
2026-11-13 15:32:58 -08:00
Rohit P 60bd6e3999 Bound the approval snippet compilation to its section 2026-11-13 15:21:46 -08:00
Rohit P b3799eed13 Fix guide example compilation and cancellation race 2026-11-13 15:16:44 -08:00
Rohit P cf2c079072 Wait for the initial UI session check before sign-in 2026-11-13 15:07:58 -08:00
Rohit P f1daf75245 Merge verified SDK fixes into cluster qualification 2026-11-13 15:07:58 -08:00
Rohit Prasad ce3bb66bae Merge pull request #897 from andrewyng/fix/service-attachment-admission
Fix attachments on service clusters
2026-11-13 14:54:27 -08:00
Rohit Prasad 63b21f9722 Merge pull request #891 from andrewyng/fix/sdk-approval-visibility
Finish SDK approvals and embedded startup fixes
2026-11-13 14:50:42 -08:00
Rohit P 38df9f7e4a Add Desktop App docs 2026-11-13 14:49:14 -08:00
Rohit P 4a34f789b0 Integrate cancellation cleanup and approval fixes 2026-11-13 14:20:41 -08:00
Rohit P 087be5a9c6 Integrate embedded helper qualification fixes 2026-11-13 14:20:17 -08:00
Rohit P 6877196b11 Merge main and preserve combined SDK release behavior 2026-11-13 14:19:56 -08:00
Rohit Prasad 8c99b01e1d Merge pull request #893 from andrewyng/feat/owcluster-cli
Rename the cluster CLI to owcluster
2026-11-13 14:17:04 -08:00
Rohit Prasad bf754ba57d Merge pull request #890 from andrewyng/fix/sdk-activity-metadata
Show tool names and arguments in SDK activity
2026-11-13 14:15:26 -08:00
Rohit Prasad 098bcbddb3 Merge pull request #889 from andrewyng/fix/sdk-release-qualification
Explain SDK agent registration conflicts
2026-11-13 14:14:22 -08:00
Rohit P be7f6f5b81 fix: inspect service attachments without a worker lease 2026-11-13 14:04:07 -08:00
Rohit P c851e6101b fix: accept DNS case variation in private cluster tunnel 2026-11-13 13:48:26 -08:00
Rohit P a67cfad969 fix: reconcile completed AWS infrastructure before installation 2026-11-13 13:39:48 -08:00
Rohit P 7390f061a5 test: report runtime causes in assistant regressions 2026-11-13 13:25:23 -08:00
Rohit P c6797d584a fix: preserve deployment review prices across restarts 2026-11-13 13:01:45 -08:00
Rohit P 6f8eec295f test: keep unrelated GCP fixtures unchanged 2026-11-13 12:41:09 -08:00
Rohit P 0c79d4864f Ignore local example credentials and signing files 2026-11-13 12:40:31 -08:00
Rohit P 95d175a2c7 fix: preserve private installation access and recording schema setup 2026-11-13 12:38:44 -08:00
Rohit P 7a0ea3c203 Fix private Azure image publication across Docker image stores 2026-11-13 12:38:44 -08:00
Rohit P d47ceb69ef Support reviewed OpenShell 0.1.8 deployments 2026-11-13 12:38:44 -08:00
Rohit P 9cd45522aa Simplify deployment configuration and qualify local Kubernetes 2026-11-13 12:38:44 -08:00
Rohit P 9fda56b234 Add Azure AWS and local cluster configuration examples 2026-11-13 12:38:44 -08:00
Rohit P abcbeadc84 Add CLI controls for management UI and trace storage 2026-11-13 12:38:44 -08:00
Rohit P febd5c038b test: keep legacy runtime runner paths nonoptional 2026-11-13 12:25:59 -08:00
Rohit P cdaa290c83 fix: handle optional runners in examples and partner packages 2026-11-13 11:58:17 -08:00
Rohit P 1ccb483807 fix: wait for cancellation cleanup before surfacing interruptions 2026-11-13 11:48:08 -08:00
Rohit P 59a17fc9a5 feat: use owcluster with a compatible ows alias 2026-11-13 11:38:12 -08:00