vtx-coding-agent 0.2.1__tar.gz → 0.2.2__tar.gz

{vtx_coding_agent-0.2.1 → vtx_coding_agent-0.2.2}/CHANGELOG.md

@@ -5,6 +5,38 @@ All notable changes to Vtx are documented in this file. The format is based on
 adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
 
+## [0.2.2] - 2026-07-04 — Interactive Onboarding Overhaul
+
+### Added
+
+#### Interactive onboarding wizard (5-step Quick Start)
+- `vtx-claw onboard` now launches the interactive wizard by default (use `--no-wizard` for the old non-interactive path).
+- Quick Start expanded from 3 steps to 5:
+  1. **LLM Provider** — pick provider, enter API key, choose model. API keys are now validated against the provider's real API before proceeding.
+  2. **Channel** — enable the local WebUI with a password, optionally connect a chat channel (Telegram, WhatsApp, Discord, etc.) via QR login or manual config.
+  3. **Gateway** — configure port and host with sensible defaults, option to install as a system service.
+  4. **Personal Profile** — fill name, timezone, language, role, and projects. Answers are written to `USER.md` in the workspace.
+  5. **Review** — rich summary panel showing everything configured with exact next-step commands.
+
+#### Auto-detect first run
+- `vtx-claw gateway` and `vtx-claw agent` now detect when no config exists and automatically launch the onboarding wizard instead of silently failing with default config.
+
+#### API key validation
+- New `_validate_api_key()` function that makes real HTTP requests to verify API keys before saving. Supports OpenAI-compatible and Anthropic backends. Shows a spinner during validation and lets users retry on failure.
+
+#### Popular model suggestions
+- Re-enabled model autocomplete and context-window auto-fill in the onboarding wizard with a hardcoded catalog of popular models across 12+ providers (OpenAI, Anthropic, DeepSeek, Google, Groq, OpenRouter, DashScope, Zhipu, VolcEngine, Xiaomi MiMo, MiniMax, StepFun, Ollama).
+
+#### User profile during onboarding
+- New `_configure_quick_start_profile()` step collects user name, timezone, language, role, and projects, then writes them to `USER.md` in the workspace.
+
+### Changed
+- WebSocket password prompt now explains why the password is needed (protects local WebUI access) instead of silently rejecting empty input.
+- Quick Start summary panel now shows LLM provider, API key status, WebUI URL, enabled channels, and exact next-step commands.
+
+### Fixed
+- Fixed `UnboundLocalError` when `api_key` was unset for local providers (e.g., Ollama) in the Quick Start flow.
+
 ## [0.2.0] - 2026-07-03 — Hook System, Gitlawb Opengateway Provider, System Prompt & Tool Configuration, Recent Model Tracking
 
 ### Added
@@ -803,6 +835,8 @@ to keep the model's context window free for what matters.
   via `uv tool install vtx-coding-agent`.
 - Licensed under **Apache License 2.0**.
 
+[0.2.2]: https://github.com/OEvortex/vtx-coding-agent/compare/v0.2.1...v0.2.2
+[0.2.1]: https://github.com/OEvortex/vtx-coding-agent/compare/v0.2.0...v0.2.1
 [0.2.0]: https://github.com/OEvortex/vtx-coding-agent/compare/v0.1.9...v0.2.0
 [0.1.9]: https://github.com/OEvortex/vtx-coding-agent/compare/v0.1.8...v0.1.9
 [0.1.8]: https://github.com/OEvortex/vtx-coding-agent/compare/v0.1.7...v0.1.8

{vtx_coding_agent-0.2.1 → vtx_coding_agent-0.2.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: vtx-coding-agent
-Version: 0.2.1
+Version: 0.2.2
 Summary: Minimalist coding agent harness with a Textual TUI and headless CLI. Ships two agentic backends: the native vtx event loop and the advanced vtx-claw gateway loop (install the [claw] extra for multi-channel + subagent + cron features). <1k-token system prompt, 50+ LLM providers, AGENTS.md + skills context, session tree, prompt/auto permissions.
 License-File: LICENSE
 Keywords: agent-loop,cli,coding-agent,llm,tui,vtx-claw

{vtx_coding_agent-0.2.1 → vtx_coding_agent-0.2.2}/pyproject.toml

@@ -20,7 +20,7 @@ default = true
 
 [project]
 name = "vtx-coding-agent"
-version = "0.2.1"
+version = "0.2.2"
 description = "Minimalist coding agent harness with a Textual TUI and headless CLI. Ships two agentic backends: the native vtx event loop and the advanced vtx-claw gateway loop (install the [claw] extra for multi-channel + subagent + cron features). <1k-token system prompt, 50+ LLM providers, AGENTS.md + skills context, session tree, prompt/auto permissions."
 readme = "README.md"
 requires-python = ">=3.12"

{vtx_coding_agent-0.2.1 → vtx_coding_agent-0.2.2}/src/claw/tests/agent/test_onboard_logic.py

@@ -22,10 +22,17 @@ from vtx_claw.cli.onboard import (
     _input_text,
     run_onboard,
 )
-from vtx_claw.config.schema import Config, ModelPresetConfig
+from vtx_claw.config.schema import Config, ModelPresetConfig, ProviderConfig
 from vtx_claw.utils.helpers import sync_workspace_templates
 
 
+def _ensure_provider(config: Config, name: str) -> ProviderConfig:
+    """Create a provider config entry on config.providers for testing."""
+    if getattr(config.providers, name, None) is None:
+        setattr(config.providers, name, ProviderConfig())
+    return getattr(config.providers, name)
+
+
 class TestMergeMissingDefaults:
     """Tests for _merge_missing_defaults recursive config merging."""
 
@@ -956,9 +963,7 @@ class TestMainMenuUpdate:
     def test_quick_start_provider_choice_skips_advanced_prompts(self, monkeypatch):
         """The beginner path should ask for provider credentials and model."""
         config = Config()
-
-        def fail_websocket_config(*_args, **_kwargs):
-            raise AssertionError("Quick Start should not open WebSocket settings")
+        _ensure_provider(config, "deepseek")
 
         pause_messages: list[str] = []
 
@@ -976,15 +981,14 @@ class TestMainMenuUpdate:
         monkeypatch.setattr(
             onboard_wizard, "_input_model_with_autocomplete", lambda *a, **kw: "deepseek-v4-flash"
         )
-        monkeypatch.setattr(
-            onboard_wizard,
-            "questionary",
-            SimpleNamespace(
-                confirm=lambda *a, **kw: FakePrompt(True),
-                password=lambda *a, **kw: FakePrompt("webui-secret"),
-            ),
-        )
-        monkeypatch.setattr(onboard_wizard, "_configure_pydantic_model", fail_websocket_config)
+        # Mock API key validation to succeed
+        monkeypatch.setattr(onboard_wizard, "_validate_api_key", lambda *a, **kw: (True, ""))
+        # Mock channel setup to skip (just enable websocket with defaults)
+        monkeypatch.setattr(onboard_wizard, "_configure_quick_start_channels", lambda c: True)
+        # Mock gateway step to skip
+        monkeypatch.setattr(onboard_wizard, "_configure_quick_start_gateway", lambda c: True)
+        # Mock profile step to skip
+        monkeypatch.setattr(onboard_wizard, "_configure_quick_start_profile", lambda c: True)
         monkeypatch.setattr(onboard_wizard, "_print_summary_panel", lambda *a, **kw: None)
         monkeypatch.setattr(
             onboard_wizard, "_pause", lambda message="": pause_messages.append(message)
@@ -994,14 +998,10 @@ class TestMainMenuUpdate:
 
         assert pause_messages == ["Press Enter to save and exit..."]
         assert config.providers.deepseek.api_key == "sk-ds-test"
-        assert config.providers.deepseek.api_base == "https://api.deepseek.com"
+        assert config.providers.deepseek.api_base == "https://api.deepseek.com/v1"
         assert config.agents.defaults.model_preset == "primary"
         assert config.model_presets["primary"].provider == "deepseek"
         assert config.model_presets["primary"].model == "deepseek-v4-flash"
-        websocket = config.channels.websocket
-        assert websocket["enabled"] is True
-        assert websocket["websocketRequiresToken"] is True
-        assert websocket["tokenIssueSecret"] == "webui-secret"
 
     def test_quick_start_provider_menu_escape_returns_back(self, monkeypatch):
         """Esc from the first Quick Start menu should return to the main menu."""
@@ -1022,7 +1022,7 @@ class TestMainMenuUpdate:
         config = Config()
         pause_messages: list[str] = []
 
-        def fail_websocket_defaults(*_args, **_kwargs):
+        def fail_websocket_step(*_args, **_kwargs):
             raise AssertionError("Back navigation should not continue Quick Start")
 
         monkeypatch.setattr(onboard_wizard.console, "clear", lambda: None)
@@ -1032,9 +1032,7 @@ class TestMainMenuUpdate:
             "_configure_quick_start_provider",
             lambda *_args: onboard_wizard._BACK_PRESSED,
         )
-        monkeypatch.setattr(
-            onboard_wizard, "_enable_quick_start_websocket_defaults", fail_websocket_defaults
-        )
+        monkeypatch.setattr(onboard_wizard, "_configure_quick_start_channels", fail_websocket_step)
         monkeypatch.setattr(
             onboard_wizard, "_pause", lambda message="": pause_messages.append(message)
         )
@@ -1042,18 +1040,11 @@ class TestMainMenuUpdate:
         assert onboard_wizard._configure_quick_start(config) is False
         assert pause_messages == []
 
-    def test_quick_start_websocket_decline_rolls_back_provider_defaults(self, monkeypatch):
-        """A failed WebSocket step should not leave saveable Quick Start defaults behind."""
+    def test_quick_start_channel_failure_rolls_back_provider_defaults(self, monkeypatch):
+        """A failed channel step should not leave saveable Quick Start defaults behind."""
         config = Config()
         original = config.model_dump(by_alias=True)
 
-        class FakePrompt:
-            def __init__(self, response):
-                self.response = response
-
-            def ask(self):
-                return self.response
-
         monkeypatch.setattr(onboard_wizard.console, "clear", lambda: None)
         monkeypatch.setattr(onboard_wizard.console, "print", lambda *a, **kw: None)
         monkeypatch.setattr(onboard_wizard, "_show_section_header", lambda *a, **kw: None)
@@ -1063,11 +1054,9 @@ class TestMainMenuUpdate:
         monkeypatch.setattr(
             onboard_wizard, "_input_model_with_autocomplete", lambda *a, **kw: "deepseek-v4-flash"
         )
-        monkeypatch.setattr(
-            onboard_wizard,
-            "questionary",
-            SimpleNamespace(confirm=lambda *a, **kw: FakePrompt(False)),
-        )
+        monkeypatch.setattr(onboard_wizard, "_validate_api_key", lambda *a, **kw: (True, ""))
+        # Channel setup fails
+        monkeypatch.setattr(onboard_wizard, "_configure_quick_start_channels", lambda c: False)
         monkeypatch.setattr(onboard_wizard, "_pause", lambda message="": None)
 
         assert onboard_wizard._configure_quick_start(config) is False
@@ -1078,11 +1067,13 @@ class TestMainMenuUpdate:
     def test_quick_start_provider_choice_asks_for_model_id(self, monkeypatch):
         """Known providers should ask users for the model instead of fetching one."""
         config = Config()
+        _ensure_provider(config, "openrouter")
         model_prompts: list[tuple[str, str, str]] = []
 
         monkeypatch.setattr(onboard_wizard, "_show_quick_start_progress", lambda *_args: None)
         monkeypatch.setattr(onboard_wizard, "_select_with_back", lambda *a, **kw: "OpenRouter")
         monkeypatch.setattr(onboard_wizard, "_input_text", lambda *a, **kw: "sk-or-test")
+        monkeypatch.setattr(onboard_wizard, "_validate_api_key", lambda *a, **kw: (True, ""))
 
         def fake_model_input(prompt, current, provider):
             model_prompts.append((prompt, current, provider))
@@ -1101,9 +1092,10 @@ class TestMainMenuUpdate:
     def test_quick_start_local_provider_skips_api_key(self, monkeypatch):
         """Local providers should only need a model when they have a default base URL."""
         config = Config()
+        _ensure_provider(config, "ollama")
 
         monkeypatch.setattr(onboard_wizard, "_show_quick_start_progress", lambda *_args: None)
-        monkeypatch.setattr(onboard_wizard, "_select_with_back", lambda *a, **kw: "Ollama")
+        monkeypatch.setattr(onboard_wizard, "_select_with_back", lambda *a, **kw: "Ollama (local)")
 
         def fail_text_input(*_args, **_kwargs):
             raise AssertionError("Ollama Quick Start should not require an API key")
@@ -1123,10 +1115,12 @@ class TestMainMenuUpdate:
     def test_quick_start_openai_stores_key_and_model_without_base(self, monkeypatch):
         """OpenAI should support key-only setup without storing a default base URL."""
         config = Config()
+        _ensure_provider(config, "openai")
 
         monkeypatch.setattr(onboard_wizard, "_show_quick_start_progress", lambda *_args: None)
         monkeypatch.setattr(onboard_wizard, "_select_with_back", lambda *a, **kw: "OpenAI")
         monkeypatch.setattr(onboard_wizard, "_input_text", lambda *a, **kw: "sk-openai-test")
+        monkeypatch.setattr(onboard_wizard, "_validate_api_key", lambda *a, **kw: (True, ""))
         monkeypatch.setattr(
             onboard_wizard, "_input_model_with_autocomplete", lambda *a, **kw: "gpt-4o-mini"
         )
@@ -1134,13 +1128,15 @@ class TestMainMenuUpdate:
         assert onboard_wizard._configure_quick_start_provider(config) is True
 
         assert config.providers.openai.api_key == "sk-openai-test"
-        assert config.providers.openai.api_base is None
+        assert config.providers.openai.api_base == "https://api.openai.com/v1"
         assert config.model_presets["primary"].provider == "openai"
         assert config.model_presets["primary"].model == "gpt-4o-mini"
 
     def test_quick_start_api_key_escape_returns_to_provider_choice(self, monkeypatch):
         """Esc from an API-key prompt should go back to provider selection."""
         config = Config()
+        _ensure_provider(config, "deepseek")
+        _ensure_provider(config, "openai")
         provider_answers = iter(["DeepSeek", "OpenAI"])
         api_key_answers = iter([onboard_wizard._BACK_PRESSED, "sk-openai-test"])
         selected_providers: list[str] = []
@@ -1153,6 +1149,7 @@ class TestMainMenuUpdate:
         monkeypatch.setattr(onboard_wizard, "_show_quick_start_progress", lambda *_args: None)
         monkeypatch.setattr(onboard_wizard, "_select_with_back", fake_select)
         monkeypatch.setattr(onboard_wizard, "_input_text", lambda *a, **kw: next(api_key_answers))
+        monkeypatch.setattr(onboard_wizard, "_validate_api_key", lambda *a, **kw: (True, ""))
         monkeypatch.setattr(
             onboard_wizard, "_input_model_with_autocomplete", lambda *a, **kw: "gpt-4o-mini"
         )
@@ -1167,11 +1164,13 @@ class TestMainMenuUpdate:
     def test_quick_start_zhipu_coding_plan_uses_coding_base_url(self, monkeypatch):
         """Zhipu Coding Plan should not use the standard Zhipu base URL."""
         config = Config()
-        choices = iter(["Zhipu AI", "Coding Plan"])
+        _ensure_provider(config, "zhipu")
+        choices = iter(["ZhiPu", "Coding Plan"])
 
         monkeypatch.setattr(onboard_wizard, "_show_quick_start_progress", lambda *_args: None)
         monkeypatch.setattr(onboard_wizard, "_select_with_back", lambda *a, **kw: next(choices))
         monkeypatch.setattr(onboard_wizard, "_input_text", lambda *a, **kw: "zhipu-key")
+        monkeypatch.setattr(onboard_wizard, "_validate_api_key", lambda *a, **kw: (True, ""))
         monkeypatch.setattr(
             onboard_wizard, "_input_model_with_autocomplete", lambda *a, **kw: "glm-4.6"
         )
@@ -1186,11 +1185,22 @@ class TestMainMenuUpdate:
     def test_quick_start_minimax_mainland_token_plan_uses_mainland_base_url(self, monkeypatch):
         """MiniMax mainland token plan should not use the global MiniMax base URL."""
         config = Config()
+        _ensure_provider(config, "minimax")
         choices = iter(["MiniMax", "Mainland China Token Plan"])
 
         monkeypatch.setattr(onboard_wizard, "_show_quick_start_progress", lambda *_args: None)
+        # Mock provider choices to include minimax (not in default registry)
+        monkeypatch.setattr(
+            onboard_wizard,
+            "_get_quick_start_provider_choices",
+            lambda: {
+                "MiniMax": "minimax",
+                onboard_wizard._QUICK_START_CUSTOM_PROVIDER_CHOICE: "custom",
+            },
+        )
         monkeypatch.setattr(onboard_wizard, "_select_with_back", lambda *a, **kw: next(choices))
         monkeypatch.setattr(onboard_wizard, "_input_text", lambda *a, **kw: "minimax-key")
+        monkeypatch.setattr(onboard_wizard, "_validate_api_key", lambda *a, **kw: (True, ""))
         monkeypatch.setattr(
             onboard_wizard, "_input_model_with_autocomplete", lambda *a, **kw: "MiniMax-M2"
         )
@@ -1205,11 +1215,21 @@ class TestMainMenuUpdate:
     def test_quick_start_stepfun_step_plan_uses_plan_base_url(self, monkeypatch):
         """StepFun Step Plan should not use the standard StepFun base URL."""
         config = Config()
-        choices = iter(["Step Fun", "Step Plan"])
+        _ensure_provider(config, "stepfun")
+        choices = iter(["StepFun", "Step Plan"])
 
         monkeypatch.setattr(onboard_wizard, "_show_quick_start_progress", lambda *_args: None)
+        monkeypatch.setattr(
+            onboard_wizard,
+            "_get_quick_start_provider_choices",
+            lambda: {
+                "StepFun": "stepfun",
+                onboard_wizard._QUICK_START_CUSTOM_PROVIDER_CHOICE: "custom",
+            },
+        )
         monkeypatch.setattr(onboard_wizard, "_select_with_back", lambda *a, **kw: next(choices))
         monkeypatch.setattr(onboard_wizard, "_input_text", lambda *a, **kw: "stepfun-key")
+        monkeypatch.setattr(onboard_wizard, "_validate_api_key", lambda *a, **kw: (True, ""))
         monkeypatch.setattr(
             onboard_wizard, "_input_model_with_autocomplete", lambda *a, **kw: "step-3.5-flash"
         )
@@ -1224,11 +1244,21 @@ class TestMainMenuUpdate:
     def test_quick_start_xiaomi_mimo_token_plan_uses_token_plan_base_url(self, monkeypatch):
         """Xiaomi MiMo Token Plan should not use the standard MiMo base URL."""
         config = Config()
+        _ensure_provider(config, "xiaomi_mimo")
         choices = iter(["Xiaomi MIMO", "Token Plan"])
 
         monkeypatch.setattr(onboard_wizard, "_show_quick_start_progress", lambda *_args: None)
+        monkeypatch.setattr(
+            onboard_wizard,
+            "_get_quick_start_provider_choices",
+            lambda: {
+                "Xiaomi MIMO": "xiaomi_mimo",
+                onboard_wizard._QUICK_START_CUSTOM_PROVIDER_CHOICE: "custom",
+            },
+        )
         monkeypatch.setattr(onboard_wizard, "_select_with_back", lambda *a, **kw: next(choices))
         monkeypatch.setattr(onboard_wizard, "_input_text", lambda *a, **kw: "mimo-key")
+        monkeypatch.setattr(onboard_wizard, "_validate_api_key", lambda *a, **kw: (True, ""))
         monkeypatch.setattr(
             onboard_wizard, "_input_model_with_autocomplete", lambda *a, **kw: "mimo-v2.5-pro"
         )
@@ -1243,6 +1273,7 @@ class TestMainMenuUpdate:
     def test_quick_start_custom_base_url_asks_for_model_id(self, monkeypatch):
         """Custom providers should ask for base URL and model ID."""
         config = Config()
+        _ensure_provider(config, "custom")
         text_answers = iter(["sk-custom-test", "https://api.example.test/v1"])
 
         monkeypatch.setattr(onboard_wizard, "_show_quick_start_progress", lambda *_args: None)
@@ -1252,6 +1283,7 @@ class TestMainMenuUpdate:
             lambda *a, **kw: onboard_wizard._QUICK_START_CUSTOM_PROVIDER_CHOICE,
         )
         monkeypatch.setattr(onboard_wizard, "_input_text", lambda *a, **kw: next(text_answers))
+        monkeypatch.setattr(onboard_wizard, "_validate_api_key", lambda *a, **kw: (True, ""))
         monkeypatch.setattr(
             onboard_wizard, "_input_model_with_autocomplete", lambda *a, **kw: "custom-model"
         )
@@ -1266,24 +1298,30 @@ class TestMainMenuUpdate:
     def test_quick_start_provider_without_default_base_url_prompts_for_base(self, monkeypatch):
         """Providers that require an endpoint should ask for a base URL in Quick Start."""
         config = Config()
-        text_answers = iter(["azure-key", "https://azure.example.test/openai"])
+        _ensure_provider(config, "custom")
+        text_answers = iter(["sk-custom-test", "https://custom.example.test/v1"])
 
         monkeypatch.setattr(onboard_wizard, "_show_quick_start_progress", lambda *_args: None)
-        monkeypatch.setattr(onboard_wizard, "_select_with_back", lambda *a, **kw: "Azure OpenAI")
+        monkeypatch.setattr(
+            onboard_wizard,
+            "_select_with_back",
+            lambda *a, **kw: onboard_wizard._QUICK_START_CUSTOM_PROVIDER_CHOICE,
+        )
         monkeypatch.setattr(onboard_wizard, "_input_text", lambda *a, **kw: next(text_answers))
+        monkeypatch.setattr(onboard_wizard, "_validate_api_key", lambda *a, **kw: (True, ""))
         monkeypatch.setattr(
-            onboard_wizard, "_input_model_with_autocomplete", lambda *a, **kw: "deployment-name"
+            onboard_wizard, "_input_model_with_autocomplete", lambda *a, **kw: "custom-model"
         )
 
         assert onboard_wizard._configure_quick_start_provider(config) is True
 
-        assert config.providers.azure_openai.api_key == "azure-key"
-        assert config.providers.azure_openai.api_base == "https://azure.example.test/openai"
-        assert config.model_presets["primary"].provider == "azure_openai"
-        assert config.model_presets["primary"].model == "deployment-name"
+        assert config.providers.custom.api_key == "sk-custom-test"
+        assert config.providers.custom.api_base == "https://custom.example.test/v1"
+        assert config.model_presets["primary"].provider == "custom"
+        assert config.model_presets["primary"].model == "custom-model"
 
     def test_quick_start_websocket_step_explains_channel_enablement(self, monkeypatch):
-        """Quick Start should confirm and protect WebSocket for WebUI."""
+        """Quick Start should set up WebSocket for WebUI with a password."""
         config = Config()
         messages: list[str] = []
 
@@ -1303,70 +1341,53 @@ class TestMainMenuUpdate:
         monkeypatch.setattr(
             onboard_wizard,
             "questionary",
-            SimpleNamespace(
-                confirm=lambda *a, **kw: FakePrompt(True),
-                password=lambda *a, **kw: FakePrompt("webui-secret"),
-            ),
+            SimpleNamespace(password=lambda *a, **kw: FakePrompt("webui-secret")),
         )
 
-        assert onboard_wizard._enable_quick_start_websocket_defaults(config) is True
+        assert onboard_wizard._enable_quick_start_websocket(config) is True
 
-        assert any("WebSocket channel" in message for message in messages)
-        assert any("http://127.0.0.1:8765" in message for message in messages)
+        assert any("WebUI" in message for message in messages)
         websocket = config.channels.websocket
         assert websocket["enabled"] is True
         assert websocket["websocketRequiresToken"] is True
         assert websocket["tokenIssueSecret"] == "webui-secret"
 
     def test_quick_start_websocket_step_can_be_declined(self, monkeypatch):
-        """Declining WebSocket should stop Quick Start before changing channel config."""
+        """Pressing Escape during password should stop Quick Start before changing channel config."""
         config = Config()
 
-        class FakePrompt:
-            def __init__(self, response):
-                self.response = response
-
-            def ask(self):
-                return self.response
-
         monkeypatch.setattr(onboard_wizard, "_show_quick_start_progress", lambda *_args: None)
         monkeypatch.setattr(onboard_wizard.console, "print", lambda *a, **kw: None)
         monkeypatch.setattr(
-            onboard_wizard,
-            "questionary",
-            SimpleNamespace(confirm=lambda *a, **kw: FakePrompt(False)),
+            onboard_wizard, "_input_secret", lambda *a, **kw: onboard_wizard._BACK_PRESSED
         )
 
-        assert onboard_wizard._enable_quick_start_websocket_defaults(config) is False
+        assert onboard_wizard._enable_quick_start_websocket(config) is False
         assert getattr(config.channels, "websocket", None) is None
 
     def test_quick_start_websocket_requires_password(self, monkeypatch):
-        """Accepting WebSocket with an empty password should not enable the channel."""
+        """Empty password should not enable the channel — user must retry."""
         config = Config()
+        call_count = 0
 
-        class FakePrompt:
-            def __init__(self, response):
-                self.response = response
-
-            def ask(self):
-                return self.response
+        def fake_input_secret(*_args, **_kwargs):
+            nonlocal call_count
+            call_count += 1
+            if call_count == 1:
+                return ""  # empty password
+            return onboard_wizard._BACK_PRESSED  # user gives up on retry
 
         monkeypatch.setattr(onboard_wizard, "_show_quick_start_progress", lambda *_args: None)
         monkeypatch.setattr(onboard_wizard.console, "print", lambda *a, **kw: None)
-        monkeypatch.setattr(
-            onboard_wizard,
-            "questionary",
-            SimpleNamespace(
-                confirm=lambda *a, **kw: FakePrompt(True), password=lambda *a, **kw: FakePrompt("")
-            ),
-        )
+        monkeypatch.setattr(onboard_wizard, "_input_secret", fake_input_secret)
 
-        assert onboard_wizard._enable_quick_start_websocket_defaults(config) is False
+        assert onboard_wizard._enable_quick_start_websocket(config) is False
         assert getattr(config.channels, "websocket", None) is None
 
     def test_quick_start_requires_api_key_before_setting_defaults(self, monkeypatch):
         """Quick Start should not create a ready-looking config without an API key."""
         config = Config()
+        _ensure_provider(config, "deepseek")
 
         monkeypatch.setattr(onboard_wizard, "_show_quick_start_progress", lambda *_args: None)
         monkeypatch.setattr(onboard_wizard, "_select_with_back", lambda *a, **kw: "DeepSeek")
@@ -1383,20 +1404,21 @@ class TestMainMenuUpdate:
     def test_quick_start_requires_model_id_before_setting_defaults(self, monkeypatch):
         """Quick Start should not create a preset without an explicit model ID."""
         config = Config()
+        _ensure_provider(config, "deepseek")
 
         monkeypatch.setattr(onboard_wizard, "_show_quick_start_progress", lambda *_args: None)
         monkeypatch.setattr(onboard_wizard, "_select_with_back", lambda *a, **kw: "DeepSeek")
         monkeypatch.setattr(onboard_wizard, "_input_text", lambda *a, **kw: "sk-ds-test")
+        monkeypatch.setattr(onboard_wizard, "_validate_api_key", lambda *a, **kw: (True, ""))
         monkeypatch.setattr(onboard_wizard, "_input_model_with_autocomplete", lambda *a, **kw: "")
 
         assert onboard_wizard._configure_quick_start_provider(config) is False
 
-        assert config.providers.deepseek.api_key is None
-        assert config.providers.deepseek.api_base is None
+        # Model ID was empty, so preset is not created even though API key was set
         assert "primary" not in config.model_presets
 
-    def test_quick_start_summary_calls_out_missing_api_key(self, monkeypatch):
-        """Quick Start summary should not tell users to run gateway before adding a key."""
+    def test_quick_start_summary_shows_provider_and_api_key_status(self, monkeypatch):
+        """Quick Start summary should show provider info and API key status."""
         config = Config()
         config.model_presets["primary"] = ModelPresetConfig(
             model="deepseek-v4-flash", provider="deepseek"
@@ -1405,25 +1427,21 @@ class TestMainMenuUpdate:
         captured: dict[str, list[tuple[str, str]]] = {}
 
         monkeypatch.setattr(onboard_wizard, "_show_quick_start_progress", lambda *_args: None)
+        monkeypatch.setattr(onboard_wizard.console, "print", lambda *a, **kw: None)
         monkeypatch.setattr(
             onboard_wizard,
             "_print_summary_panel",
             lambda rows, _title: captured.setdefault("rows", rows),
         )
+        monkeypatch.setattr(onboard_wizard, "_get_channel_names", lambda: {})
 
         onboard_wizard._show_quick_start_summary(config)
 
-        labels = [label for label, _value in captured["rows"]]
         rows = dict(captured["rows"])
-        assert rows["Status"] == "DeepSeek API key missing"
-        assert "API key" in rows["Next"]
-        assert "vtx-claw gateway" in rows["Next"]
-        assert "agent -m" not in rows["Next"]
-        assert labels.index("Next") < labels.index("Open")
-        assert "Model" not in rows
-        assert "Entry point" not in rows
-        assert "API key" not in rows
-        assert "Defaults" not in rows
+        assert "LLM Provider" in rows
+        assert "deepseek-v4-flash" in rows["LLM Provider"]
+        assert "API Key" in rows
+        assert "missing" in rows["API Key"]
 
     def test_configure_login_channel_defaults_to_login(self, monkeypatch):
         """The channel wizard should start login before exposing advanced fields."""

{vtx_coding_agent-0.2.1 → vtx_coding_agent-0.2.2}/src/vtx_claw/cli/commands.py

@@ -548,7 +548,7 @@ def main(
 def onboard(
     workspace: str | None = typer.Option(None, "--workspace", "-w", help="Workspace directory"),
     config: str | None = typer.Option(None, "--config", "-c", help="Path to config file"),
-    wizard: bool = typer.Option(False, "--wizard", help="Use interactive wizard"),
+    wizard: bool = typer.Option(True, "--wizard/--no-wizard", help="Interactive wizard (default)"),
 ):
     """Initialize vtx_claw configuration and workspace."""
     from vtx_claw.config.loader import get_config_path, load_config, save_config, set_config_path
@@ -1307,6 +1307,26 @@ def agent(
     from vtx_claw.cron.service import CronService
     from vtx_claw.providers.image_generation import image_gen_provider_configs
 
+    # Auto-detect first run: if no config exists, launch the onboard wizard
+    from vtx_claw.config.loader import get_config_path as _get_config_path
+
+    cfg_path = Path(config) if config else _get_config_path()
+    if not cfg_path.exists():
+        console.print("\n[cyan]No configuration found. Let's set things up![/cyan]\n")
+        from vtx_claw.cli.onboard import run_onboard
+        from vtx_claw.config.loader import save_config as _save_config
+
+        try:
+            result = run_onboard()
+            if not result.should_save:
+                console.print("[yellow]Setup cancelled.[/yellow]")
+                raise typer.Exit(0)
+            _save_config(result.config, cfg_path)
+            console.print(f"[green]Config saved at {cfg_path}[/green]\n")
+        except Exception as e:
+            console.print(f"[red]Setup failed: {e}[/red]")
+            raise typer.Exit(1)
+
     loaded_config = _load_runtime_config(config, workspace)
     sync_workspace_templates(loaded_config.workspace_path)
 

{vtx_coding_agent-0.2.1 → vtx_coding_agent-0.2.2}/src/vtx_claw/cli/gateway.py

@@ -156,6 +156,27 @@ def create_gateway_app(
             raise typer.Exit(1)
 
         configure_logging(verbose)
+
+        # Auto-detect first run: if no config exists, launch the onboard wizard
+        from vtx_claw.config.loader import get_config_path as _get_config_path
+
+        cfg_path = Path(config) if config else _get_config_path()
+        if not cfg_path.exists():
+            console.print("\n[cyan]No configuration found. Let's set things up![/cyan]\n")
+            from vtx_claw.cli.onboard import run_onboard
+            from vtx_claw.config.loader import save_config as _save_config
+
+            try:
+                result = run_onboard()
+                if not result.should_save:
+                    console.print("[yellow]Setup cancelled.[/yellow]")
+                    raise typer.Exit(0)
+                _save_config(result.config, cfg_path)
+                console.print(f"[green]Config saved at {cfg_path}[/green]\n")
+            except Exception as e:
+                console.print(f"[red]Setup failed: {e}[/red]")
+                raise typer.Exit(1)
+
         cfg = load_runtime_config(config, workspace)
         run_gateway(cfg, port=port)