kolega-code 0.2.0__tar.gz → 0.3.0__tar.gz

{kolega_code-0.2.0 → kolega_code-0.3.0}/.env.example

@@ -6,6 +6,8 @@ OPENAI_API_KEY=
 GOOGLE_API_KEY=
 MOONSHOT_API_KEY=
 DEEPSEEK_API_KEY=
+ZAI_API_KEY=
+KIMI_CODING_API_KEY=
 
 KOLEGA_CODE_THINKING_EFFORT=
 

{kolega_code-0.2.0 → kolega_code-0.3.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: kolega-code
-Version: 0.2.0
+Version: 0.3.0
 Summary: Local-first AI coding agent for the terminal
 Project-URL: Homepage, https://kolega-ai.github.io/kolega-code/
 Project-URL: Documentation, https://kolega-ai.github.io/kolega-code/
@@ -49,6 +49,7 @@ Requires-Dist: jinja2==3.1.6
 Requires-Dist: langfuse==3.14.5
 Requires-Dist: nest-asyncio==1.6.0
 Requires-Dist: openai==1.109.1
+Requires-Dist: packaging==25.0
 Requires-Dist: pathspec==1.1.1
 Requires-Dist: playwright==1.59.0
 Requires-Dist: psutil==7.2.2
@@ -96,10 +97,13 @@ kolega-code --version
 Upgrade or uninstall:
 
 ```bash
-uv tool upgrade kolega-code
+kolega-code update
 uv tool uninstall kolega-code
 ```
 
+Running the installer again also updates an existing install to the latest
+released version.
+
 Run the Textual UI and open the Settings tab to pick any provider and model from the catalog, choose the model's thinking effort, and save your API key:
 
 ```bash

{kolega_code-0.2.0 → kolega_code-0.3.0}/README.md

@@ -28,10 +28,13 @@ kolega-code --version
 Upgrade or uninstall:
 
 ```bash
-uv tool upgrade kolega-code
+kolega-code update
 uv tool uninstall kolega-code
 ```
 
+Running the installer again also updates an existing install to the latest
+released version.
+
 Run the Textual UI and open the Settings tab to pick any provider and model from the catalog, choose the model's thinking effort, and save your API key:
 
 ```bash

{kolega_code-0.2.0 → kolega_code-0.3.0}/RELEASING.md

@@ -29,8 +29,8 @@ the organization request.
 4. Create and push a matching tag:
 
    ```bash
-   git tag v0.2.0
-   git push origin v0.2.0
+   git tag v0.3.0
+   git push origin v0.3.0
    ```
 
 5. Approve the `pypi` environment deployment in GitHub Actions.

{kolega_code-0.2.0 → kolega_code-0.3.0}/docs/src/content/docs/cli/doctor.md

@@ -33,6 +33,8 @@ you can validate a specific provider/model combination before using it.
   itself is never printed).
 - **Configuration** — whether a valid `AgentConfig` can be built. If valid, it
   prints the resolved **long**, **fast**, and **thinking** models.
+- **Update** — whether the installed CLI is current or a newer PyPI release is
+  available.
 
 ## Example output
 
@@ -40,6 +42,7 @@ you can validate a specific provider/model combination before using it.
 Project: /Users/you/code/my-app
 State dir: /Users/you/Library/Application Support/kolega-code
 Textual installed: True
+Update: Kolega Code is up to date (0.3.0).
 Stored active model: moonshot/kimi-k2.7-code
 Stored thinking effort: auto
 Stored API key: present in local settings

{kolega_code-0.2.0 → kolega_code-0.3.0}/docs/src/content/docs/cli/overview.md

@@ -6,7 +6,7 @@ description: Invocation forms, global options, and the command surface of kolega
 The `kolega-code` command has two shapes:
 
 - **No subcommand** → launches the interactive [Terminal UI](../../tui/interface/).
-- **A subcommand** (`ask`, `sessions`, `doctor`) → runs a specific non-interactive
+- **A subcommand** (`ask`, `sessions`, `doctor`, `update`) → runs a specific non-interactive
   task.
 
 ```bash
@@ -14,6 +14,7 @@ kolega-code [PROJECT_PATH] [options]      # interactive TUI
 kolega-code ask "<prompt>" [options]       # one-shot prompt
 kolega-code sessions <list|delete|export> [options]
 kolega-code doctor [options]
+kolega-code update
 ```
 
 ## Commands
@@ -24,6 +25,7 @@ kolega-code doctor [options]
 | [`ask`](../ask/) | Run a single prompt and print the answer |
 | [`sessions`](../sessions/) | List, delete, or export saved sessions |
 | [`doctor`](../doctor/) | Check local configuration and API-key status |
+| `update` | Update Kolega Code to the latest released version |
 
 ## Launching the TUI
 

{kolega_code-0.2.0 → kolega_code-0.3.0}/docs/src/content/docs/configuration/environment-variables.md

@@ -44,8 +44,14 @@ provider or model.
 | `DASHSCOPE_API_KEY` | DashScope (Alibaba) |
 | `MOONSHOT_API_KEY` | Moonshot |
 | `DEEPSEEK_API_KEY` | DeepSeek |
-
-The local `llama` provider needs no key.
+| `ZAI_API_KEY` | Z.AI (GLM Coding Plan) |
+| `KIMI_CODING_API_KEY` | Kimi Coding Plan |
+
+The local `llama` provider needs no key. The `zai` key authenticates against Z.AI's
+Anthropic-compatible endpoint (it is the key Z.AI also documents as `ANTHROPIC_AUTH_TOKEN`).
+The `kimi_coding` key authenticates against the Kimi Coding Plan's separate
+Anthropic-compatible endpoint (`https://api.kimi.com/coding/`), which is distinct from the
+standard Moonshot API used by the `moonshot` provider.
 
 ## Model selection
 

{kolega_code-0.2.0 → kolega_code-0.3.0}/docs/src/content/docs/configuration/providers-and-models.mdx

@@ -24,8 +24,26 @@ The agent runtime recognizes these providers:
 | DashScope (Alibaba) | `dashscope` | `DASHSCOPE_API_KEY` |
 | Moonshot (Kimi) | `moonshot` | `MOONSHOT_API_KEY` |
 | DeepSeek | `deepseek` | `DEEPSEEK_API_KEY` |
+| Z.AI (GLM Coding Plan) | `zai` | `ZAI_API_KEY` |
+| Kimi Coding Plan | `kimi_coding` | `KIMI_CODING_API_KEY` |
 | Llama (local) | `llama` | _none — local model_ |
 
+<Aside type="tip">
+**Z.AI GLM Coding Plan.** The `zai` provider targets Z.AI's Anthropic-compatible
+endpoint (`https://api.z.ai/api/anthropic`) and exposes **GLM-5.2** (default) and
+**GLM-5.1**. Set `ZAI_API_KEY` to your GLM Coding Plan key (the key Z.AI also
+documents as `ANTHROPIC_AUTH_TOKEN`).
+</Aside>
+
+<Aside type="tip">
+**Kimi Coding Plan.** The `kimi_coding` provider targets Kimi's Anthropic-compatible
+coding endpoint (`https://api.kimi.com/coding/`), which is separate from the standard
+Moonshot API used by the `moonshot` provider. It exposes a single stable model ID,
+**`kimi-for-coding`**, that the backend auto-routes to the latest underlying model:
+with thinking effort `auto` (enabled) requests route to **K2.7 Code**, and with `none`
+(disabled) they route to **K2.6**. Set `KIMI_CODING_API_KEY` to your Kimi Coding Plan key.
+</Aside>
+
 <Aside type="note">
 The **Settings tab** in the TUI currently offers a curated shortlist —
 **Moonshot (Kimi K2.7 Code and Kimi K2.6)** and **DeepSeek (DeepSeek V4 Pro)**.
@@ -96,6 +114,8 @@ models in the TUI.
 | DeepSeek `deepseek-v4-pro` | `none`, `high`, `max` | `high` |
 | Moonshot `kimi-k2.7-code` | `auto` | `auto` |
 | Moonshot `kimi-k2.6` | `auto`, `none` | `auto` |
+| Z.AI `glm-5.2` | `high`, `max` | `max` |
+| Z.AI `glm-5.1` | `auto`, `none` | `auto` |
 | Google `gemini-2.5-pro` | `auto`, `low`, `medium`, `high` | `medium` |
 | OpenAI reasoning models | `low`, `medium`, `high` | `medium` |
 

{kolega_code-0.2.0 → kolega_code-0.3.0}/docs/src/content/docs/getting-started/installation.mdx

@@ -58,10 +58,17 @@ present. See [`doctor`](../../cli/doctor/) for details.
 
 ## Upgrade or uninstall
 
-If you installed with the script or `uv tool install`, manage the CLI with uv:
+If you installed with the script or `uv tool install`, update the CLI with the
+built-in command:
+
+```bash
+kolega-code update
+```
+
+Running the installer again also updates an existing install to the latest
+released version. To uninstall, use uv:
 
 ```bash
-uv tool upgrade kolega-code
 uv tool uninstall kolega-code
 ```
 

{kolega_code-0.2.0 → kolega_code-0.3.0}/docs/src/content/docs/tui/sessions-and-resume.md

@@ -10,7 +10,7 @@ task list. Sessions let you stop and resume work later.
 ## How sessions are created
 
 - **In the TUI**, launching `kolega-code .` starts a **fresh session by default**.
-  Your work is saved when you quit (`Ctrl+Q` or `/quit`).
+  Your work is saved when you quit (`Ctrl+Q`, `/quit`, or `/exit`).
 - **With `ask`**, a session is only persisted if you pass `--save` or `--session`.
   See [`kolega-code ask`](../../cli/ask/).
 

{kolega_code-0.2.0 → kolega_code-0.3.0}/docs/src/content/docs/tui/slash-commands.md

@@ -39,7 +39,9 @@ These control the app and your session.
 | `/effort` | Choose the active model's thinking effort |
 | `/copy` | Copy the last response to the clipboard |
 | `/version` | Show the Kolega Code version |
+| `/update` | Update Kolega Code to the latest version |
 | `/quit` | Save the session and exit |
+| `/exit` | Save the session and exit |
 
 Run `/model` to open a selectable list of supported models for the current
 provider. You can also switch directly with `/model <name>`.

{kolega_code-0.2.0 → kolega_code-0.3.0}/kolega_code/__init__.py

@@ -148,4 +148,4 @@ __all__ = [
     "parse_git_status_output",
 ]
 
-__version__ = "0.2.0"
+__version__ = "0.3.0"

{kolega_code-0.2.0 → kolega_code-0.3.0}/kolega_code/agent/__init__.py

@@ -39,4 +39,4 @@ __all__ = [
     "ToolExtension",
 ]
 
-__version__ = "0.2.0"
+__version__ = "0.3.0"

{kolega_code-0.2.0 → kolega_code-0.3.0}/kolega_code/agent/tests/llm/test_thinking_effort.py

@@ -17,6 +17,12 @@ def test_model_specs_expose_provider_specific_thinking_efforts() -> None:
     assert default_thinking_effort("deepseek", "deepseek-v4-pro") == "high"
     assert thinking_effort_options("google", "gemini-3.1-pro-preview") == ("low", "medium", "high")
     assert default_thinking_effort("google", "gemini-3.1-pro-preview") == "high"
+    assert thinking_effort_options("zai", "glm-5.2") == ("high", "max")
+    assert default_thinking_effort("zai", "glm-5.2") == "max"
+    assert thinking_effort_options("zai", "glm-5.1") == ("auto", "none")
+    assert default_thinking_effort("zai", "glm-5.1") == "auto"
+    assert thinking_effort_options("kimi_coding", "kimi-for-coding") == ("auto", "none")
+    assert default_thinking_effort("kimi_coding", "kimi-for-coding") == "auto"
 
 
 def test_anthropic_opus_effort_uses_adaptive_thinking_without_budget_tokens() -> None:
@@ -48,6 +54,32 @@ def test_deepseek_thinking_effort_serialization() -> None:
     assert max_params["output_config"] == {"effort": "max"}
 
 
+def test_zai_glm52_thinking_effort_serialization() -> None:
+    provider = AnthropicProvider(api_key="test-key", provider_name="zai")
+
+    max_params = {"model": "glm-5.2"}
+    provider._apply_thinking_params(max_params, GenerationParams(thinking="max"))
+    assert max_params["thinking"] == {"type": "enabled"}
+    assert max_params["output_config"] == {"effort": "max"}
+
+    high_params = {"model": "glm-5.2"}
+    provider._apply_thinking_params(high_params, GenerationParams(thinking="high"))
+    assert high_params["output_config"] == {"effort": "high"}
+
+
+def test_zai_glm51_thinking_toggle_serialization() -> None:
+    provider = AnthropicProvider(api_key="test-key", provider_name="zai")
+
+    auto_params = {"model": "glm-5.1"}
+    provider._apply_thinking_params(auto_params, GenerationParams(thinking="auto"))
+    # GLM-5.1 is a plain toggle — no output_config.effort field.
+    assert auto_params == {"model": "glm-5.1", "thinking": {"type": "enabled"}}
+
+    disabled_params = {"model": "glm-5.1"}
+    provider._apply_thinking_params(disabled_params, GenerationParams(thinking="none"))
+    assert disabled_params == {"model": "glm-5.1", "thinking": {"type": "disabled"}}
+
+
 def test_moonshot_kimi_thinking_toggle_serialization() -> None:
     provider = AnthropicProvider(api_key="test-key", provider_name="moonshot")
 
@@ -64,6 +96,18 @@ def test_moonshot_kimi_thinking_toggle_serialization() -> None:
     assert disabled_params == {"model": "kimi-k2.6", "thinking": {"type": "disabled"}}
 
 
+def test_kimi_coding_thinking_toggle_serialization() -> None:
+    provider = AnthropicProvider(api_key="test-key", provider_name="kimi_coding")
+
+    enabled = {"model": "kimi-for-coding"}
+    provider._apply_thinking_params(enabled, GenerationParams(thinking="auto"))
+    assert enabled == {"model": "kimi-for-coding", "thinking": {"type": "enabled"}}
+
+    disabled = {"model": "kimi-for-coding"}
+    provider._apply_thinking_params(disabled, GenerationParams(thinking="none"))
+    assert disabled == {"model": "kimi-for-coding", "thinking": {"type": "disabled"}}
+
+
 def test_google_gemini_3_pro_uses_thinking_level() -> None:
     provider = GoogleProvider(api_key="test-key")
 

{kolega_code-0.2.0 → kolega_code-0.3.0}/kolega_code/cli/app.py

@@ -47,7 +47,6 @@ from textual.widgets import (
 )
 from textual.widgets.option_list import Option
 
-from kolega_code import __version__ as kolega_code_version
 from kolega_code.agent import AgentConfig, AgentEvent, CoderAgent, PlanningAgent, PromptExtension, ToolExtension
 from kolega_code.llm.exceptions import LLMError, llm_error_message
 from kolega_code.llm.models import Message, MessageHistory, TextBlock, ToolCall, ToolResult
@@ -81,6 +80,7 @@ from .slash_commands import (
     agent_command_names,
     search_commands,
 )
+from .updater import check_for_update, run_self_update, update_status_message
 from .skills import (
     SkillCatalog,
     activated_skill_names,
@@ -719,6 +719,7 @@ class KolegaCodeApp(App):
         settings_store: Optional[SettingsStore] = None,
         overrides: Optional[CliConfigOverrides] = None,
         browser_visible: bool = False,
+        check_for_updates: bool = False,
     ) -> None:
         super().__init__()
         self.project_path = project_path
@@ -735,6 +736,7 @@ class KolegaCodeApp(App):
         self.skill_catalog: SkillCatalog = discover_skills(self.project_path)
         self.file_index = WorkspaceFileIndex(self.project_path)
         self.browser_visible = browser_visible
+        self.check_for_updates = check_for_updates
         self.connection_manager = CliConnectionManager()
         self.agent: Optional[CoderAgent | PlanningAgent] = None
         self.agent_worker = None
@@ -842,6 +844,8 @@ class KolegaCodeApp(App):
         self._set_effort_actions_visible(False)
         self._refresh_planning_sidebar()
         self._ensure_startup_entry()
+        if self.check_for_updates:
+            self.run_worker(self._check_for_update_on_startup(), name="kolega-update-check", group="updates")
         self._conversation.anchor()
         self.run_worker(self._consume_events(), name="kolega-events", group="events")
         if self.config is not None:
@@ -942,6 +946,14 @@ class KolegaCodeApp(App):
         except Exception:
             pass
 
+    async def _check_for_update_on_startup(self) -> None:
+        result = await asyncio.to_thread(check_for_update)
+        message = update_status_message(result)
+        if not message:
+            return
+        self._add_conversation_entry(ConversationEntry(kind="system", content=message))
+        self._notify_user(message)
+
     @property
     def _status_dashboard(self) -> Static:
         return self.query_one("#status_dashboard", Static)
@@ -1614,7 +1626,9 @@ class KolegaCodeApp(App):
             "/effort": self._command_effort,
             "/copy": self._command_copy,
             "/version": self._command_version,
+            "/update": self._command_update,
             "/quit": self._command_quit,
+            "/exit": self._command_quit,
         }
 
     async def _handle_tui_slash_command(self, stripped_text: str, composer: ChatComposer) -> bool:
@@ -1857,10 +1871,41 @@ class KolegaCodeApp(App):
         self._notify_user(messages.COPY_LAST_RESPONSE)
 
     async def _command_version(self, args: str) -> None:
+        result = await asyncio.to_thread(check_for_update)
+        lines = [messages.VERSION_INFO.format(version=result.current_version)]
+        update_message = update_status_message(result, include_up_to_date=True, include_errors=True)
+        if update_message:
+            lines.append(update_message)
         self._add_conversation_entry(
-            ConversationEntry(kind="system", content=messages.VERSION_INFO.format(version=kolega_code_version))
+            ConversationEntry(kind="system", content="\n".join(lines))
         )
 
+    async def _command_update(self, args: str) -> None:
+        if self._turn_active or self.agent_worker is not None:
+            self._show_composer_hint(messages.BLOCK_STOP_BEFORE_UPDATE)
+            self._notify_user(messages.BLOCK_STOP_BEFORE_UPDATE, severity="warning")
+            return
+
+        self._notify_user(messages.UPDATE_STARTED)
+        result = await asyncio.to_thread(run_self_update, capture_output=True)
+        severity = "information" if result.returncode == 0 else "error"
+        if result.returncode == 0:
+            lines = [messages.UPDATE_COMPLETED]
+        else:
+            lines = [messages.UPDATE_FAILED.format(code=result.returncode)]
+            if result.error:
+                lines.append(result.error)
+
+        output = "\n".join(part for part in (result.stdout.strip(), result.stderr.strip()) if part)
+        if output:
+            if len(output) > 4000:
+                output = "[output truncated]\n" + output[-4000:]
+            lines.extend(["", "Output:", "```text", output, "```"])
+
+        content = "\n".join(lines)
+        self._add_conversation_entry(ConversationEntry(kind="system", content=content))
+        self._notify_user(lines[0], severity=severity)
+
     async def _command_quit(self, args: str) -> None:
         await self.action_quit()
 

{kolega_code-0.2.0 → kolega_code-0.3.0}/kolega_code/cli/config.py

@@ -41,6 +41,8 @@ API_KEY_ENV = {
     ModelProvider.DASHSCOPE: "DASHSCOPE_API_KEY",
     ModelProvider.MOONSHOT: "MOONSHOT_API_KEY",
     ModelProvider.DEEPSEEK: "DEEPSEEK_API_KEY",
+    ModelProvider.ZAI: "ZAI_API_KEY",
+    ModelProvider.KIMI_CODING: "KIMI_CODING_API_KEY",
 }
 
 
@@ -255,6 +257,8 @@ def build_agent_config(
             dashscope_api_key=_api_key_for_provider(ModelProvider.DASHSCOPE, loaded_env, settings),
             moonshot_api_key=_api_key_for_provider(ModelProvider.MOONSHOT, loaded_env, settings),
             deepseek_api_key=_api_key_for_provider(ModelProvider.DEEPSEEK, loaded_env, settings),
+            zai_api_key=_api_key_for_provider(ModelProvider.ZAI, loaded_env, settings),
+            kimi_coding_api_key=_api_key_for_provider(ModelProvider.KIMI_CODING, loaded_env, settings),
             environment=overrides.environment or loaded_env.get("KOLEGA_CODE_ENVIRONMENT", "development"),
             long_context_config=_model_config(long_provider, long_model, thinking_effort=active_thinking_effort),
             fast_config=_model_config(fast_provider, fast_model),

{kolega_code-0.2.0 → kolega_code-0.3.0}/kolega_code/cli/main.py

@@ -10,7 +10,6 @@ import sys
 from pathlib import Path
 from typing import Iterable, Optional
 
-from kolega_code import __version__
 from kolega_code.agent import CoderAgent
 from kolega_code.llm.exceptions import LLMBillingError, billing_error_message
 from kolega_code.llm.models import TextBlock
@@ -37,8 +36,9 @@ from .skills import (
     build_skill_tool_extension,
     discover_skills,
 )
+from .updater import check_for_update, run_self_update, update_status_message
 
-SUBCOMMANDS = {"ask", "sessions", "doctor"}
+SUBCOMMANDS = {"ask", "sessions", "doctor", "update"}
 RESUME_LATEST = "__latest__"
 CLI_AGENT_MODE = AgentMode.CLI.value
 
@@ -46,12 +46,16 @@ CLI_AGENT_MODE = AgentMode.CLI.value
 def main(argv: Optional[Iterable[str]] = None) -> int:
     args = parse_args(list(argv) if argv is not None else sys.argv[1:])
     try:
+        if getattr(args, "version", False):
+            return _run_version()
         if args.command == "ask":
             return asyncio.run(_run_ask(args))
         if args.command == "sessions":
             return _run_sessions(args)
         if args.command == "doctor":
             return _run_doctor(args)
+        if args.command == "update":
+            return _run_update()
         return _run_tui(args)
     except (CliConfigError, SessionStoreError, SettingsStoreError, ValueError) as exc:
         _print_styled(f"kolega-code: {exc}", style="error", stderr=True)
@@ -110,7 +114,7 @@ def _add_session_args(parser: argparse.ArgumentParser, session_help: str = "Sess
 def _build_tui_parser() -> argparse.ArgumentParser:
     parser = argparse.ArgumentParser(prog="kolega-code", description="Run the Kolega Code Textual CLI.")
     parser.set_defaults(command="tui")
-    parser.add_argument("--version", action="version", version=f"%(prog)s {__version__}")
+    parser.add_argument("--version", action="store_true", help="Show the Kolega Code version.")
     parser.add_argument("project_path", nargs="?", default=".", type=Path, help="Project directory to work in.")
     parser.add_argument("--mode", choices=[mode.value for mode in AgentMode], default=CLI_AGENT_MODE, help=argparse.SUPPRESS)
     parser.add_argument("--new", action="store_true", help="Start a new session. This is now the default.")
@@ -159,6 +163,8 @@ def _build_subcommand_parser() -> argparse.ArgumentParser:
     doctor.add_argument("--state-dir", type=Path, help="Directory for CLI session state.")
     _add_common_model_args(doctor)
 
+    subparsers.add_parser("update", help="Update Kolega Code to the latest version.")
+
     return parser
 
 
@@ -260,6 +266,26 @@ def _resolve_tui_session(
     return store.create(project_path, CLI_AGENT_MODE, summary)
 
 
+def _run_version() -> int:
+    result = check_for_update()
+    print(f"kolega-code {result.current_version}")
+    message = update_status_message(result)
+    if message:
+        print(message)
+    return 0
+
+
+def _run_update() -> int:
+    result = run_self_update()
+    if result.error:
+        _print_styled(result.error, style="error", stderr=True)
+    if result.returncode == 0:
+        print("Kolega Code update completed. Run `kolega-code --version` to confirm.")
+    elif not result.error:
+        _print_styled("Kolega Code update failed.", style="error", stderr=True)
+    return result.returncode
+
+
 def _run_tui(args: argparse.Namespace) -> int:
     if importlib.util.find_spec("textual") is None:
         print("Textual is not installed. Reinstall the CLI with: uv tool install --force kolega-code", file=sys.stderr)
@@ -296,6 +322,7 @@ def _run_tui(args: argparse.Namespace) -> int:
         overrides=_overrides_from_args(args),
         session=session,
         browser_visible=args.browser_visible,
+        check_for_updates=True,
     )
     app.run()
     return 0
@@ -571,6 +598,9 @@ def _run_doctor(args: argparse.Namespace) -> int:
     line("State dir", store.root)
     textual_installed = importlib.util.find_spec("textual") is not None
     line("Textual installed", textual_installed, "success" if textual_installed else "warning")
+    update_message = update_status_message(check_for_update(), include_up_to_date=True, include_errors=True)
+    if update_message:
+        line("Update", update_message)
     if settings.active_provider and settings.active_model:
         line("Stored active model", f"{settings.active_provider}/{settings.active_model}")
         line("Stored thinking effort", settings.active_thinking_effort or "model default")

{kolega_code-0.2.0 → kolega_code-0.3.0}/kolega_code/cli/messages.py

@@ -65,6 +65,9 @@ EFFORT_SWITCH_HINT = "Choose below, or switch with /effort <level>."
 COPY_LAST_RESPONSE = "Copied the last response to the clipboard."
 COPY_NOTHING = "No response to copy yet."
 VERSION_INFO = "Kolega Code version {version}."
+UPDATE_STARTED = "Updating Kolega Code…"
+UPDATE_COMPLETED = "Kolega Code update completed. Restart this TUI to use the updated version."
+UPDATE_FAILED = "Kolega Code update failed with exit code {code}."
 
 # Blockers
 BLOCK_STOP_BEFORE_RESET = "Stop the current turn before resetting the thread."
@@ -72,6 +75,7 @@ BLOCK_STOP_BEFORE_MODE_SWITCH = "Stop the current turn before switching modes."
 BLOCK_STOP_BEFORE_SKILL = "Stop the current turn before activating a skill."
 BLOCK_STOP_BEFORE_MODEL_SWITCH = "Stop the current turn before switching models."
 BLOCK_STOP_BEFORE_EFFORT_SWITCH = "Stop the current turn before switching thinking effort."
+BLOCK_STOP_BEFORE_UPDATE = "Stop the current turn before updating Kolega Code."
 BLOCK_PLAN_DECISION = "Choose Implement plan or Discuss further before sending another message."
 BLOCK_PLAN_DECISION_MODE_SWITCH = "Choose Implement plan or Discuss further before switching modes."
 BLOCK_PLAN_DECISION_SKILL = "Choose Implement plan or Discuss further before activating a skill."

{kolega_code-0.2.0 → kolega_code-0.3.0}/kolega_code/cli/provider_registry.py

@@ -24,6 +24,8 @@ from kolega_code.llm.specs import (
 PROVIDER_LABELS: dict[ModelProvider, str] = {
     ModelProvider.MOONSHOT: "Moonshot AI",
     ModelProvider.DEEPSEEK: "DeepSeek AI",
+    ModelProvider.ZAI: "Z.AI (GLM Coding Plan)",
+    ModelProvider.KIMI_CODING: "Kimi Coding Plan",
     ModelProvider.ANTHROPIC: "Anthropic",
     ModelProvider.OPENAI: "OpenAI",
     ModelProvider.GOOGLE: "Google",
@@ -40,9 +42,14 @@ MODEL_LABELS: dict[str, str] = {
     "kimi-k2.7-code": "Kimi K2.7 Code",
     "kimi-k2.7-code-highspeed": "Kimi K2.7 Code (High-Speed)",
     "kimi-k2.6": "Kimi K2.6",
+    # Kimi Coding Plan
+    "kimi-for-coding": "Kimi for Coding",
     # DeepSeek
     "deepseek-v4-pro": "DeepSeek V4 Pro",
     "deepseek-v4-flash": "DeepSeek V4 Flash",
+    # Z.AI (GLM Coding Plan)
+    "glm-5.2": "GLM-5.2",
+    "glm-5.1": "GLM-5.1",
     # Anthropic
     "claude-opus-4-8": "Claude Opus 4.8",
     "claude-opus-4-7": "Claude Opus 4.7",
@@ -76,6 +83,8 @@ MODEL_LABELS: dict[str, str] = {
 PROVIDER_DEFAULT_MODEL: dict[ModelProvider, str] = {
     ModelProvider.MOONSHOT: "kimi-k2.7-code",
     ModelProvider.DEEPSEEK: "deepseek-v4-pro",
+    ModelProvider.ZAI: "glm-5.2",
+    ModelProvider.KIMI_CODING: "kimi-for-coding",
     ModelProvider.ANTHROPIC: "claude-opus-4-8",
     ModelProvider.OPENAI: "gpt-5.5",
     ModelProvider.GOOGLE: "gemini-3.1-pro-preview",

{kolega_code-0.2.0 → kolega_code-0.3.0}/kolega_code/cli/slash_commands.py

@@ -46,7 +46,9 @@ TUI_COMMAND_ENTRIES: tuple[SlashCommandEntry, ...] = (
     SlashCommandEntry("effort", "Show or set the active thinking effort", CommandScope.TUI),
     SlashCommandEntry("copy", "Copy the last response to the clipboard", CommandScope.TUI),
     SlashCommandEntry("version", "Show the Kolega Code version", CommandScope.TUI),
+    SlashCommandEntry("update", "Update Kolega Code to the latest version", CommandScope.TUI),
     SlashCommandEntry("quit", "Save the session and exit", CommandScope.TUI),
+    SlashCommandEntry("exit", "Save the session and exit", CommandScope.TUI),
 )
 
 TUI_COMMAND_NAMES: frozenset[str] = frozenset(entry.token for entry in TUI_COMMAND_ENTRIES)