zames_pro 2.72.9 → 2.73.1

package/CHANGELOG.md

@@ -7,6 +7,85 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## [Unreleased]
 
+## [2.73.1] - 2026-10-11
+
+### Fixed
+
+- **qwen:** live provider switch, verified send, server-busy handling
+
+## [2.73.0] - 2026-10-11
+
+### Added
+
+- **qwen:** a second chat provider (`chat.qwen.ai`), selectable with
+  `/provider <deepseek|qwen>`, so the operator can switch models without
+  leaving zames. Adds `src/qwen-ui.ts` (selectors), `src/qwen-net-capture.ts`
+  (phase-aware SSE parser), `src/qwen-browser.ts` (`QwenBrowser`), a separate
+  persistent profile (`~/.zames/qwen-profile`), `/model` to list/pick a Qwen
+  model, and file attachments via the mode menu's "Upload attachment" item.
+- **provider:** switching `/provider` mid-session now carries the conversation
+  over: the current chat is summarized (on the old backend) and persisted as a
+  handoff (`src/provider-handoff.ts`); the next launch posts it as the first
+  message of the new provider's chat, so context is not lost. Falls back to a
+  local history dump when the model summary is unusable. Unit-tested.
+- **provider:** a `PROVIDERS` registry (`src/providers.ts`) is the SINGLE
+  source of truth (ids, labels, `/chat` URL prefix), so adding a backend
+  (ChatGPT, Gemini, …) is a data change plus the provider's own modules — not a
+  dozen ternaries. `/provider` accepts any registered id; the config enum, the
+  `{provider}` label and `/doctor` read from the registry. Unit-tested
+  (`test/providers.test.ts`).
+- **provider:** a `ChatProvider` interface (`src/types.ts`) both backends
+  implement; the agent loop only depends on `BrowserLike` and never sees a
+  concrete class. Operator notices (`ds.answer_timeout`, `ds.tool_retry`, …)
+  now use a `{provider}` label so they never say "DeepSeek" while driving Qwen.
+
+### Changed
+
+- **ui:** the status line shows a provider badge ([Qwen]/[DeepSeek]) so the
+  active backend is always visible in a multi-provider session.
+- **qwen:** `listChats` / `fetchChatMessages` now use Qwen's OWN API
+  (`GET /api/v2/chats/?page=1&exclude_project=true` and
+  `GET /api/v2/chats/<id>?direction=up&limit=N`) with the DOM scrape as a
+  fallback, so `/chats` and `/resume` work on a signed-in account (verified
+  live: 9 chats with titles, 20 messages). Parsers are pure + tested.
+- **qwen:** password login (`_autoLogin`) fixed to the verified order (email ->
+  "Log in with a password" -> password -> submit); the email-CODE flow needs the
+  operator to read the mail, so a saved password is preferred. Verified live.
+- **qwen:** the context counter now uses `input_tokens` (the prompt size) —
+  the true "used context" — instead of `total_tokens` (which adds the answer),
+  so the status bar matches DeepSeek's semantics.
+- **qwen:** thinking / web search are mapped onto Qwen's own UI: deepThinking
+  ON selects the "Thinking" mode (OFF → "Fast") in the
+  `.qwen-thinking-selector` dropdown, and webSearch toggles the "Web search"
+  item in the "+" mode menu. `/thinking` and `/web` therefore work on Qwen too.
+- **qwen:** no 15 s send throttle — a live experiment (4 back-to-back sends at
+  0 ms gap) hit no rate limit, so `qwen.minSendIntervalMs` defaults to 1000 ms.
+  The agent-loop watchdog deadline is derived from the Qwen answer timeout
+  (a cold guest turn can take minutes), not the DeepSeek 240 s default.
+- **provider:** sessions are tagged with the ACTIVE provider
+  (`saveSession({…, provider})`), so `/resume`/`/last`/auto-restore only ever
+  reopen a chat in the backend that created it (a chat id is meaningless to the
+  other one). `loadLastSession(workdir, provider)` filters accordingly.
+- **qwen:** the restored-dialogue scraper moved to a PURE `src/qwen-chats.ts`
+  (mirroring `browser-chats.ts`) and now keeps only the OUTERMOST message
+  container — the broad `[class*="message"]` chain matched an outer and an
+  inner block, so every turn printed twice. `listChats` opens the sidebar first
+  and matches `/c/<id>` links. Compact / auto-compact use the active provider's
+  own answer timeout (`providerAnswerTimeoutMs()`), not DeepSeek's.
+
+### Fixed
+
+- **qwen:** recover from a lost page/context mid-run (relaunch + re-login, and
+  re-navigate to the chat if the input is gone) instead of failing every retry.
+- **qwen:** the guest daily limit (`getstsToken` → `RateLimited`) fails fast
+  with a "sign in" message instead of a 20-minute rate-limit wait.
+- **qwen:** the real sign-in flow is **email → one-time code from the letter**
+  (not email+password): `_promptAndLogin()` now sends the email, asks the
+  operator for the code in the terminal, submits it, and verifies the session
+  from the page (guest header CTAs absent). `/provider login` re-runs it.
+- **qwen:** a longer send-button wait when files are attached (the upload keeps
+  the button disabled).
+
 ## [2.72.9] - 2026-10-10
 
 ### Fixed

package/README.md

@@ -5,7 +5,7 @@
 <h1 align="center">zames_pro</h1>
 
 <p align="center">
-  <strong>A terminal coding agent that drives chat.deepseek.com through Playwright — no API key required.</strong>
+  <strong>A terminal coding agent that drives real web chat LLMs (DeepSeek, Qwen, …) through a browser — no API key required.</strong>
 </p>
 
 <p align="center">
@@ -22,7 +22,9 @@
   <strong>English</strong> | <a href="#readme-ru">Русский</a>
 </p>
 
-A terminal coding agent that works on top of [chat.deepseek.com](https://chat.deepseek.com/) through Playwright.
+A terminal coding agent that works on top of real web chat LLMs
+([chat.deepseek.com](https://chat.deepseek.com/), [chat.qwen.ai](https://chat.qwen.ai/), …)
+through Playwright.
 In spirit it is similar to Claude Code / Codex CLI: it starts in the current
 directory, reads and edits files, runs commands, and commits to git.
 
@@ -58,9 +60,14 @@ and runs right after it, in the same chat. `Esc` aborts the current generation,
 ## How it works
 
 zames does not call the model API. It launches a headless Chromium with a
-persistent profile, signs in to `chat.deepseek.com` like a human, types the task
+persistent profile, signs in to a chat web app like a human, types the task
 into the chat box, and reads the answer back.
 
+Two chat backends are available, switched with `/provider <deepseek|qwen>`
+(restart required): **DeepSeek** (`chat.deepseek.com`, the default) and
+**Qwen** (`chat.qwen.ai`). Each keeps its own persistent profile, so switching
+never signs the other one out.
+
 <p align="center">
   <img src="https://raw.githubusercontent.com/Viqto0r/zames_pro/master/docs/how-it-works.png" alt="how zames works" width="860">
 </p>
@@ -84,6 +91,7 @@ Key pieces:
 
 - [Quick start](#quick-start) · [Features](#features) · [Why zames?](#why-zames) · [Requirements](#requirements)
 - [How it works](#how-it-works) · [Installation](#installation) · [Signing in](#signing-in) · [Usage](#usage)
+- [Switching providers (DeepSeek / Qwen)](#switching-providers-deepseek--qwen)
 - [Tools](#tools) · [Slash commands](#slash-commands)
 - [Project context, skills and memory](#project-context-skills-and-memory) · [MCP (external tools)](#mcp-external-tools) · [Configuration](#configuration)
 - [FAQ](#faq) · [Links](#links) · [License](#license)
@@ -122,11 +130,12 @@ Key pieces:
 ## Requirements
 
 - Node.js >= 20
-- A DeepSeek account. On first launch zames asks for your DeepSeek
-  login/password in the terminal (and stores them in `~/.zames/config.json`
-  after a successful sign-in, so a later logout is handled automatically
-  without asking you again). You can also sign in manually in the browser
-  window when the browser is headed.
+- An account on the chat you want to use — **DeepSeek** (`chat.deepseek.com`,
+  the default) or **Qwen** (`chat.qwen.ai`). Qwen also works in guest mode for
+  a few messages. On first launch zames asks for your login in the terminal
+  (and stores it in `~/.zames/config.json` after a successful sign-in, so a
+  later logout is handled automatically without asking you again). You can also
+  sign in manually in the browser window when the browser is headed.
 
 ## Signing in
 
@@ -159,6 +168,36 @@ version). You do not need `--headed` just to log in.
 Credentials and toggles can also be edited from `/config`
 (`browser.auth.username`, `browser.auth.password`, `browser.auth.saveSession`).
 
+## Switching providers (DeepSeek / Qwen)
+
+zames can drive either **DeepSeek** (`chat.deepseek.com`, the default) or
+**Qwen** (`chat.qwen.ai`):
+
+```bash
+/provider qwen      # switch to Qwen (persisted); restart zames to take effect
+/provider deepseek  # switch back
+/provider status    # show the active provider / model / login state
+/provider login     # (re)run the sign-in check
+/model              # list Qwen models
+/model qwen3.7-plus # pick one
+```
+
+Notes:
+
+- The two backends use **separate persistent profiles**
+  (`~/.zames/profile` for DeepSeek, `~/.zames/qwen-profile` for Qwen), so
+  switching never signs the other one out. A provider switch takes effect on
+  the next launch — the browser is created once at startup.
+- Qwen works in **guest mode** without an account, but a guest is limited to a
+  handful of messages per day; for long agent runs sign in (email / Google /
+  GitHub — no phone required). zames auto-logins from the saved credentials or
+  asks in the terminal, exactly like DeepSeek. `/model` changes the Qwen model
+  at runtime (no restart).
+- File attachments on Qwen go through the mode menu's **Upload attachment**
+  item; DeepSeek uses its own paperclip.
+- `/doctor` shows the active provider and the UI health of that provider's
+  page.
+
 ## Installation
 
 ```bash
@@ -302,7 +341,11 @@ Codex CLI:
   context size in tokens (DeepSeek's `accumulated_token_usage`).
 - /export [file] — write the session transcript to a Markdown file
   (zames-export-<stamp>.md by default).
-- /doctor — diagnose node, git, config, browser, clipboard and MCP.
+- /doctor — diagnose node, git, config, browser, clipboard, MCP and the
+  active chat provider (it also reports that provider's UI health).
+- /provider <deepseek|qwen> — switch the chat backend (restart to apply);
+  /provider status shows the active provider/model/login; /provider login
+  re-runs the sign-in check. /model [id] lists or picks a Qwen model at runtime.
 - /add-dir <path> — validate an extra directory (the sandbox is fixed at
   startup; relaunch with --dir to write there).
 - /resume <n> (after /chats) and /resume-id <id> — open a chat and PRINT its
@@ -329,7 +372,10 @@ tests`). /cron "<min> <hour> <dom> <month> <dow>" <task> — run on a schedule.
   /jobs [rm <id>|clear] lists and stops them. A fired job is put into the same
   message queue you type into, so it runs when the agent is free (never mid-
   generation) and still respects the send throttle.
-- /thinking [on|off] and /web [on|off] — toggle Deep thinking / Smart search.
+- /thinking [on|off] and /web [on|off] — toggle the reasoning / web-search
+  behaviour. On **DeepSeek** these are the "Deep thinking" / "Smart search"
+  toggles; on **Qwen** they drive its thinking MODE (Thinking / Fast) and the
+  "Web search" item in the "+" menu.
   These (and /queue, /jobs, /goal, `/config <sub>`) can be used WHILE the agent
   is working — they do not touch the in-flight generation.
 - /queue [clear] — list or clear the messages waiting to be sent after the
@@ -494,7 +540,9 @@ MIT
 <details>
 <summary>🇷🇺 Читать по-русски (Russian)</summary>
 
-Терминальный coding-агент, работающий поверх [chat.deepseek.com](https://chat.deepseek.com/) через Playwright.
+Терминальный coding-агент, работающий поверх веб-чатов LLM
+([chat.deepseek.com](https://chat.deepseek.com/), [chat.qwen.ai](https://chat.qwen.ai/), …)
+через Playwright.
 По духу он похож на Claude Code / Codex CLI: запускается в текущей
 директории, читает и правит файлы, выполняет команды и коммитит в git.
 
@@ -531,8 +579,13 @@ zames                      # один раз войти, дальше прост
 ## Как это работает
 
 zames не вызывает API модели. Он запускает headless Chromium с постоянным
-профилем, входит в `chat.deepseek.com` как человек, печатает задачу в поле
-чата и читает ответ обратно.
+профилем, входит в веб-чат как человек, печатает задачу в поле чата и читает
+ответ обратно.
+
+Доступны два бэкенда, переключаются командой `/provider <deepseek|qwen>`
+(нужен перезапуск): **DeepSeek** (`chat.deepseek.com`, по умолчанию) и
+**Qwen** (`chat.qwen.ai`). У каждого свой постоянный профиль, поэтому
+переключение не разлогинивает другой.
 
 <p align="center">
   <img src="https://raw.githubusercontent.com/Viqto0r/zames_pro/master/docs/how-it-works.png" alt="как работает zames" width="860">
@@ -560,6 +613,7 @@ zames не вызывает API модели. Он запускает headless C
 
 - [Быстрый старт](#быстрый-старт) · [Возможности](#возможности) · [Почему zames?](#почему-zames) · [Требования](#требования)
 - [Как это работает](#как-это-работает) · [Установка](#установка) · [Вход в аккаунт](#вход-в-аккаунт) · [Использование](#использование)
+- [Переключение провайдера (DeepSeek / Qwen)](#переключение-провайдера-deepseek--qwen)
 - [Инструменты](#инструменты) · [Слэш-команды](#слэш-команды)
 - [Контекст проекта, навыки и память](#контекст-проекта-навыки-и-память) · [MCP (внешние инструменты)](#mcp-внешние-инструменты) · [Конфигурация](#конфигурация)
 - [FAQ](#faq) · [Ссылки](#ссылки) · [Лицензия](#лицензия)
@@ -600,10 +654,12 @@ zames не вызывает API модели. Он запускает headless C
 ## Требования
 
 - Node.js >= 20
-- Аккаунт DeepSeek. При первом запуске zames спросит логин/пароль DeepSeek в
-  терминале (и сохранит их в `~/.zames/config.json` после успешного входа,
-  поэтому более поздний разлогин обрабатывается автоматически, без повторных
-  вопросов). Можно также войти вручную в окне браузера, когда он headed.
+- Аккаунт в нужном чате — **DeepSeek** (`chat.deepseek.com`, по умолчанию) или
+  **Qwen** (`chat.qwen.ai`). Qwen также работает в гостевом режиме на несколько
+  сообщений. При первом запуске zames спросит логин в терминале (и сохранит его
+  в `~/.zames/config.json` после успешного входа, поэтому более поздний
+  разлогин обрабатывается автоматически, без повторных вопросов). Можно также
+  войти вручную в окне браузера, когда он headed.
 
 ## Вход в аккаунт
 
@@ -636,6 +692,35 @@ Headless работает из коробки: headless Chrome обычно пр
 Учётные данные и переключатели можно также править из `/config`
 (`browser.auth.username`, `browser.auth.password`, `browser.auth.saveSession`).
 
+## Переключение провайдера (DeepSeek / Qwen)
+
+zames умеет работать с **DeepSeek** (`chat.deepseek.com`, по умолчанию) и
+**Qwen** (`chat.qwen.ai`):
+
+```bash
+/provider qwen      # переключиться на Qwen (сохраняется); нужен перезапуск
+/provider deepseek  # вернуться обратно
+/provider status    # активный провайдер / модель / статус входа
+/provider login     # повторно пройти проверку входа
+/model              # список моделей Qwen
+/model qwen3.7-plus # выбрать модель
+```
+
+Замечания:
+
+- У бэкендов **разные постоянные профили** (`~/.zames/profile` у DeepSeek,
+  `~/.zames/qwen-profile` у Qwen), поэтому переключение не разлогинивает
+  другой. Смена провайдера применяется при следующем запуске — браузер
+  создаётся один раз на старте.
+- Qwen работает в **гостевом режиме** без аккаунта, но гостю доступно всего
+  несколько сообщений в день; для долгих прогонов войдите (email / Google /
+  GitHub — телефон не нужен). zames сам логинится по сохранённым кредам или
+  спросит в терминале, как и с DeepSeek. `/model` меняет модель Qwen на лету
+  (без перезапуска).
+- Вложения в Qwen идут через пункт **Upload attachment** в меню режима; в
+  DeepSeek — своя скрепка.
+- `/doctor` показывает активный провайдер и состояние UI его страницы.
+
 ## Установка
 
 ```bash
@@ -785,7 +870,11 @@ zames --task "..." --output-format jsonl \
   `accumulated_token_usage`).
 - /export [file] — записать транскрипт сессии в Markdown-файл
   (по умолчанию zames-export-<stamp>.md).
-- /doctor — диагностика node, git, конфига, браузера, буфера обмена и MCP.
+- /doctor — диагностика node, git, конфига, браузера, буфера обмена, MCP и
+  активного провайдера чата (плюс состояние UI его страницы).
+- /provider <deepseek|qwen> — сменить бэкенд чата (нужен перезапуск);
+  /provider status показывает активный провайдер/модель/вход; /provider login
+  повторно проверяет вход. /model [id] покажет или выберет модель Qwen на лету.
 - /add-dir <path> — проверить дополнительную директорию (песочница
   фиксируется при старте; для записи туда перезапустите с --dir).
 - /resume <n> (после /chats) и /resume-id <id> — открыть чат и НАПЕЧАТАТЬ его
@@ -814,7 +903,10 @@ zames --task "..." --output-format jsonl \
   Сработавшая задача попадает в ту же очередь сообщений, что и ваш ввод, так
   что выполняется, когда агент свободен (никогда посреди генерации), и всё
   равно соблюдает троттлинг отправки.
-- /thinking [on|off] и /web [on|off] — переключить Deep thinking / Smart search.
+- /thinking [on|off] и /web [on|off] — переключить режим размышления /
+  веб-поиска. В **DeepSeek** это тумблеры «Deep thinking» / «Smart search»;
+  в **Qwen** — это режим размышления (Thinking / Fast) и пункт «Web search»
+  в меню «+».
   Их (а также /queue, /jobs, /goal, `/config <sub>`) можно использовать, ПОКА
   агент работает — они не трогают текущую генерацию.
 - /queue [clear] — показать или очистить сообщения, ждущие отправки после

package/dist/agent-loop.js

@@ -1,4 +1,5 @@
 import { buildSystemPrompt } from './system-prompt.js';
+import { providerLabel as providerHumanLabel } from './providers.js';
 import { loadProjectContext } from './context.js';
 import { getGitContext, formatGitContext } from './gitTools.js';
 import { translate } from './i18n.js';
@@ -52,6 +53,11 @@ export async function runAgentLoop({ browser, tools, task, workdir, maxIteration
     const safeToolResult = safe(onToolResult);
     const safeAssistantMessage = safe(onAssistantMessage);
     const safeWarning = safe(onWarning);
+    // Human label of the active backend, so provider-agnostic notices do not say
+    // "DeepSeek" while the loop is actually driving Qwen. Falls back to DeepSeek
+    // when the browser (a test double) does not expose a provider name.
+    const providerLabel = providerHumanLabel(browser.providerName || 'deepseek');
+    const tr = (key, params) => translate(locale)(key, { provider: providerLabel, ...params });
     // One AbortController per task. A tool gets `toolAbort.signal`; the loop
     // aborts it as soon as the operator presses Esc/Ctrl+C (browser._abort /
     // _stopped). This is what actually kills a running `npm test` instead of
@@ -323,7 +329,7 @@ export async function runAgentLoop({ browser, tools, task, workdir, maxIteration
             });
             // Show how many watchdog retries remain, so the operator can tell a
             // single hiccup from a genuine stall (the budget used to be invisible).
-            safeWarning(translate(locale)('ds.answer_timeout', {
+            safeWarning(tr('ds.answer_timeout', {
                 sec: Math.round(askDeadlineMs / 1000),
                 attempt: Math.min(afterToolRetries + 1, MAX_AFTER_TOOL_RETRIES),
                 max: MAX_AFTER_TOOL_RETRIES,
@@ -336,7 +342,7 @@ export async function runAgentLoop({ browser, tools, task, workdir, maxIteration
             transcript?.log('ask_timeout_exhausted', {
                 message: 'ask() did not return an answer and the retry limit is exhausted',
             });
-            safeWarning(translate(locale)('ds.answer_timeout_give_up'));
+            safeWarning(tr('ds.answer_timeout_give_up'));
             return 'ask() watchdog: no model answer received';
         }
         await reportChat();
@@ -532,7 +538,7 @@ export async function runAgentLoop({ browser, tools, task, workdir, maxIteration
                 // once per task, otherwise the retry budget spams the terminal.
                 if (!toolRetryWarned) {
                     toolRetryWarned = true;
-                    safeWarning(translate(locale)('ds.tool_retry', {
+                    safeWarning(tr('ds.tool_retry', {
                         attempt: unparsedRetries,
                         max: MAX_UNPARSED_RETRIES,
                     }));
@@ -541,9 +547,11 @@ export async function runAgentLoop({ browser, tools, task, workdir, maxIteration
                     (justRanTool
                         ? 'You stopped after a tool call and wrote plain text. '
                         : 'You wrote plain text without a tool call. ') +
+                        'Plain text is a PROTOCOL VIOLATION — the operator never sees it. ' +
                         'The task is not finished. Reply with EXACTLY one JSON tool-call ' +
                         'object, no text before or after, for example: ' +
                         '{\"tool\": \"Bash\", \"args\": {\"command\": \"...\"}}. ' +
+                        'Do NOT explain, plan or narrate — call a tool. ' +
                         'If the task is really done — call respond with the final message.';
                 continue;
             }
@@ -604,7 +612,9 @@ export async function runAgentLoop({ browser, tools, task, workdir, maxIteration
                     ? lastText +
                         String.fromCharCode(10) +
                         String.fromCharCode(10) +
-                        '(The model stopped calling tools before finishing. The task may be incomplete - check the DeepSeek chat.)'
+                        '(The model stopped calling tools before finishing. The task may be incomplete - check the ' +
+                        providerLabel +
+                        ' chat.)'
                     : 'The model stopped calling tools before finishing the task.';
             }
             const suspiciousFinal = responseLooksLikeToolCall(rawResponse) ||
@@ -676,7 +686,9 @@ export async function runAgentLoop({ browser, tools, task, workdir, maxIteration
             if (!isMeaningfulRespond(msg)) {
                 transcript?.log('empty_respond_exhausted', { response: rawResponse });
                 safeWarning('The model called respond without text, and the retry limit is exhausted. ' +
-                    'Check the DeepSeek chat manually.');
+                    'Check the ' +
+                    providerLabel +
+                    ' chat manually.');
                 return 'The model produced no final message (empty respond).';
             }
             safeAssistantMessage(msg);

package/dist/browser-pure.js

@@ -224,28 +224,28 @@ export class GenerationIncompleteError extends Error {
     }
 }
 // ---------- profile cleanup ----------
-export async function cleanSingletonFiles() {
+export async function cleanSingletonFiles(dir = USER_DATA_DIR) {
     const names = ['SingletonLock', 'SingletonCookie', 'SingletonSocket'];
     for (const name of names) {
-        const p = path.join(USER_DATA_DIR, name);
+        const p = path.join(dir, name);
         try {
             await fs.unlink(p);
         }
         catch { }
     }
 }
-export async function profileLooksLocked() {
+export async function profileLooksLocked(dir = USER_DATA_DIR) {
     try {
-        await fs.access(path.join(USER_DATA_DIR, 'SingletonLock'));
+        await fs.access(path.join(dir, 'SingletonLock'));
         return true;
     }
     catch {
         return false;
     }
 }
-export async function nukeProfile() {
+export async function nukeProfile(dir = USER_DATA_DIR) {
     try {
-        await fs.rm(USER_DATA_DIR, { recursive: true, force: true });
+        await fs.rm(dir, { recursive: true, force: true });
     }
     catch { }
 }

package/dist/browser.js

@@ -22,6 +22,7 @@ export { STOP_NAME_RE, CONTINUE_NAME_RE };
 import { askPassword, USER_DATA_DIR, parseHeadlessUACache, headlessEngineKey, readCachedHeadlessUA, writeCachedHeadlessUA, headlessUACacheReady, sanitizeHeadlessUA, AUTH_MARKER_FILE, isGenerationIncompleteText, isRateLimitText, isServerBusyText, normText, cleanAnswer, RateLimitError, ServerBusyError, GenerationIncompleteError, cleanSingletonFiles, profileLooksLocked, nukeProfile, } from './browser-pure.js';
 export { askPassword, parseHeadlessUACache, headlessUACacheReady, sanitizeHeadlessUA, isGenerationIncompleteText, isRateLimitText, isServerBusyText, normText, RateLimitError, ServerBusyError, GenerationIncompleteError, };
 export class DeepSeekBrowser {
+    providerName = 'deepseek';
     headless;
     debug;
     channel;
@@ -214,7 +215,10 @@ export class DeepSeekBrowser {
     }
     /** Translation bound to this browser's locale. */
     _t(key, params) {
-        return translate(this.locale)(key, params);
+        // Default the {provider} placeholder to DeepSeek for the shared `ds.*`
+        // messages (QwenBrowser overrides it with 'Qwen'), so a provider-agnostic
+        // string never lies about which backend is talking.
+        return translate(this.locale)(key, { provider: 'DeepSeek', ...params });
     }
     /**
      * Switch the language of the messages the browser itself prints (login

package/dist/commands.js

@@ -1,3 +1,4 @@
+import { providerLabel } from './providers.js';
 import path from 'path';
 import { isImageName } from './attachments.js';
 import { translate } from './i18n.js';
@@ -276,6 +277,10 @@ export function renderDoctor(d, t = translate('en')) {
     row(true, 'node', d.nodeVersion);
     row(true, 'platform', d.platform);
     row(true, 'workdir', d.workdir);
+    if (d.provider) {
+        const label = providerLabel(d.provider).toLowerCase();
+        row(true, 'provider', label + (d.providerModel ? ' (' + d.providerModel + ')' : ''));
+    }
     row(d.gitOk, 'git', d.gitOk
         ? t('doctor.repo', { v: d.gitBranch || t('doctor.detached') })
         : t('doctor.not_repo'));
@@ -1013,6 +1018,8 @@ export const SLASH_COMMANDS = [
     { name: '/queue', key: 'help.cmd.queue' },
     { name: '/tasks', key: 'help.cmd.tasks' },
     { name: '/config', key: 'help.cmd.config' },
+    { name: '/provider', key: 'help.cmd.provider' },
+    { name: '/model', key: 'help.cmd.model' },
     { name: '/skills', key: 'help.cmd.skills' },
     { name: '/memory', key: 'help.cmd.memory' },
     { name: '/remember', key: 'help.cmd.remember' },

package/dist/config.js

@@ -4,6 +4,7 @@ import os from 'os';
 import { DEFAULT_DIAGNOSTICS } from './diagnostics.js';
 import { DEFAULT_LOCALE } from './i18n.js';
 import { writeJsonAtomic } from './fsutil.js';
+import { PROVIDER_IDS } from './providers.js';
 const ZAMES_HOME = path.join(os.homedir(), '.zames');
 const HOME_CONFIG = path.join(ZAMES_HOME, 'config.json');
 // The PROJECT config path is resolved LAZILY: `/cd` changes process.cwd() at
@@ -14,8 +15,23 @@ export function projectConfigPath() {
 }
 export const DEFAULTS = {
     maxIterations: 0,
+    // Default backend stays DeepSeek so existing users see no change; the
+    // operator switches with /provider qwen (persisted to the home config).
+    provider: 'deepseek',
+    qwen: {
+        model: '',
+        enabled: false,
+        // Qwen tolerates back-to-back sends (verified live: 4 sends at 0 ms gap,
+        // no rate limit), so the DeepSeek 15 s throttle is pure latency here.
+        minSendIntervalMs: 1000,
+        // The first send after a cold load can take minutes ("Model loading…" +
+        // thinking); a GUEST turn measured up to ~9 min live on a cold profile, so
+        // give it generous room. A signed-in session answers in seconds; this only
+        // bounds the worst case.
+        answerTimeoutMs: 900000,
+    },
     // Headless by default: the browser runs without a visible window. Set
-    // headless: false (or pass --headed) to watch/debug the DeepSeek page.
+    // headless: false (or pass --headed) to watch/debug the page.
     headless: true,
     debug: false,
     browserChannel: null,
@@ -91,6 +107,16 @@ export function loadConfig() {
     }
     return merged;
 }
+// The agent-loop watchdog deadline per provider. The loop default (240s) is
+// tuned for DeepSeek; a Qwen GUEST turn regularly takes 7-8 minutes (cold
+// "Model loading…" + a long thinking phase) and would trip the watchdog on
+// every send. Derive the deadline from the provider's own answer timeout so
+// there is headroom. DeepSeek keeps browser.askDeadlineMs (the loop default).
+export function providerAskDeadlineMs(cfg) {
+    if (cfg.provider === 'qwen')
+        return cfg.qwen.answerTimeoutMs + 120_000;
+    return cfg.browser.askDeadlineMs;
+}
 function deepMerge(target, source) {
     for (const key of Object.keys(source)) {
         if (source[key] &&
@@ -152,6 +178,35 @@ export const CONFIG_SCHEMA = [
         labelKey: 'cfg.f.browser_authSaveSession',
         groupKey: 'cfg.group.browser',
     },
+    {
+        path: 'provider',
+        type: 'enum',
+        values: [...PROVIDER_IDS],
+        labelKey: 'cfg.f.provider',
+        groupKey: 'cfg.group.browser',
+    },
+    {
+        path: 'qwen.model',
+        type: 'string',
+        labelKey: 'cfg.f.qwen_model',
+        groupKey: 'cfg.group.browser',
+    },
+    {
+        path: 'qwen.minSendIntervalMs',
+        type: 'number',
+        min: 0,
+        max: 600000,
+        labelKey: 'cfg.f.qwen_minSendIntervalMs',
+        groupKey: 'cfg.group.browser',
+    },
+    {
+        path: 'qwen.answerTimeoutMs',
+        type: 'number',
+        min: 1000,
+        max: 3600000,
+        labelKey: 'cfg.f.qwen_answerTimeoutMs',
+        groupKey: 'cfg.group.browser',
+    },
     {
         path: 'debug',
         type: 'boolean',

package/dist/i18n/common.js

@@ -1,8 +1,8 @@
 // Common UI strings: app/common, status, tasks, messages, diff, context, undo, cd.
 export const commonMessages = {
     'app.tagline': {
-        ru: 'агент поверх chat.deepseek.com через Playwright',
-        en: 'coding agent over chat.deepseek.com via Playwright',
+        ru: 'агент поверх chat.deepseek.com / chat.qwen.ai через Playwright',
+        en: 'coding agent over chat.deepseek.com / chat.qwen.ai via Playwright',
     },
     'common.yes': { ru: 'да', en: 'yes' },
     'common.no': { ru: 'нет', en: 'no' },

package/dist/i18n/config.js

@@ -6,6 +6,20 @@ export const configMessages = {
     'cfg.group.checkpoint': { ru: 'Чекпойнты', en: 'Checkpoints' },
     'cfg.group.transcript': { ru: 'Транскрипт', en: 'Transcript' },
     'cfg.group.browser': { ru: 'Браузер / DeepSeek', en: 'Browser / DeepSeek' },
+    'cfg.group.qwen': { ru: 'Qwen', en: 'Qwen' },
+    'cfg.f.provider': { ru: 'Провайдер чата', en: 'Chat provider' },
+    'cfg.f.qwen_model': {
+        ru: 'Модель Qwen (id)',
+        en: 'Qwen model (id)',
+    },
+    'cfg.f.qwen_minSendIntervalMs': {
+        ru: 'Qwen: пауза между отправками (мс)',
+        en: 'Qwen: pause between sends (ms)',
+    },
+    'cfg.f.qwen_answerTimeoutMs': {
+        ru: 'Qwen: таймаут ответа (мс)',
+        en: 'Qwen: answer timeout (ms)',
+    },
     'cfg.f.ui_contextLimit': {
         ru: 'Размер контекста (токены)',
         en: 'Context window (tokens)',