pythinker-code 0.58.0__tar.gz → 0.59.0__tar.gz

{pythinker_code-0.58.0 → pythinker_code-0.59.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: pythinker-code
-Version: 0.58.0
+Version: 0.59.0
 Summary: Pythinker — an agentic CLI developed by Pythoughts-labs.
 Keywords: cli,agent,ai,coding-assistant,llm,claude,openai,terminal
 Author: Pythoughts-labs
@@ -115,13 +115,13 @@ It speaks the [**Agent Client Protocol (ACP)**](https://github.com/agentclientpr
 
 ---
 
-## 🆕 What's New in 0.58.0
+## 🆕 What's New in 0.59.0
 
-- **Agent core deepened with an observable request-assembly path.** Provider handoff, prompt ordering, and context persistence now run through explicit compatibility contracts, conversation history updates are transactional (atomic replacement with cancellation and rollback), and agent/skill definitions resolve through one source-aware catalogue instead of scattered lookups.
-- **Agent-spec loading is more defensive and truthful.** Subagent `path`, `extend`, and `system_prompt_path` references that resolve outside their spec's directory are now rejected instead of loaded, and malformed frontmatter parsing errors are no longer misreported as harmless skips.
-- **Thinking and subagent activity render cleanly in the terminal.** Live reasoning previews render complete Markdown without exposing stray HTML comments, and activity-tree rows stay visually stable during long-running turns.
+- **Explicit Z.AI compatibility routes.** Independent Z.AI Coding Plan and API login routes now use separate credentials, endpoints, model identities, catalog refresh, logout, and usage/rate-limit state, with immutable compatibility profiles keeping request-format quirks behind the chat-provider boundary.
+- **Streamed tool calls are correlated safely.** Interleaved argument chunks stay attached to their indexed calls, malformed or truncated call streams stop before tool execution, and failed attempts are not retried after output has already been shown.
+- **Deterministic reviewer scopes and supervised tool batches.** Reviewer subagents receive fully resolved Git commit anchors with invalid or empty scopes rejected explicitly, and tool execution runs through a supervised batch engine with ordered results, bounded cancellation, and fail-closed new batches until cleanup drains.
 
-Upgrade with `pythinker update`, `pip install --upgrade pythinker-code==0.58.0`, or use the native installer for your platform from the [Releases page](https://github.com/Pythoughts-labs/pythinker-code/releases/latest).
+Upgrade with `pythinker update`, `pip install --upgrade pythinker-code==0.59.0`, or use the native installer for your platform from the [Releases page](https://github.com/Pythoughts-labs/pythinker-code/releases/latest).
 
 
 ---
@@ -227,7 +227,7 @@ matches your OS — no Python, Node, or `uv` prerequisite.
 
 | Platform | Recommended install | Artifact source |
 |---|---|---|
-| **🪟 Windows** | `irm https://pythinker.com/install.ps1 \| iex` | `PythinkerSetup-0.58.0.exe` from [Releases](https://github.com/Pythoughts-labs/pythinker-code/releases/latest) |
+| **🪟 Windows** | `irm https://pythinker.com/install.ps1 \| iex` | `PythinkerSetup-0.59.0.exe` from [Releases](https://github.com/Pythoughts-labs/pythinker-code/releases/latest) |
 | **<img src="https://img.shields.io/badge/-macOS-000000?style=flat-square&logo=apple&logoColor=white" alt="macOS"> / <img src="https://img.shields.io/badge/-Linux-FCC624?style=flat-square&logo=linux&logoColor=black" alt="Linux">** | `curl -fsSL https://pythinker.com/install.sh \| bash` | native tarball from [Releases](https://github.com/Pythoughts-labs/pythinker-code/releases/latest) |
 | **<img src="https://img.shields.io/badge/-macOS-000000?style=flat-square&logo=apple&logoColor=white" alt="macOS"> — Homebrew** | `brew install Pythoughts-labs/pythinker/pythinker-code` | auto-published Homebrew tap |
 | **🐳 Docker** | `docker run --rm -it ghcr.io/pythoughts-labs/pythinker-code` | GHCR multi-arch image |
@@ -255,7 +255,7 @@ pythinker                      # start the interactive TUI
 
 ### 🪟 Windows — native installer
 
-`PythinkerSetup-0.58.0.exe` is an Inno Setup wizard. Release builds are signed
+`PythinkerSetup-0.59.0.exe` is an Inno Setup wizard. Release builds are signed
 when Authenticode secrets are configured in CI; otherwise the installer ships
 unsigned. Installs per-user into `%LOCALAPPDATA%\Programs\Pythinker`, registers
 `pythinker` on your user PATH (`HKCU\Environment`), broadcasts
@@ -267,13 +267,13 @@ irm https://pythinker.com/install.ps1 | iex
 
 # Or manually download the installer + checksum from the Releases page,
 # verify with Get-FileHash, then run:
-.\PythinkerSetup-0.58.0.exe
+.\PythinkerSetup-0.59.0.exe
 
 # Open a fresh PowerShell
 pythinker --version
 ```
 
-**Per-machine install** (IT-managed boxes): `.\PythinkerSetup-0.58.0.exe /ALLUSERS`
+**Per-machine install** (IT-managed boxes): `.\PythinkerSetup-0.59.0.exe /ALLUSERS`
 installs to `%ProgramFiles%\Pythinker` and writes PATH to HKLM (requires admin).
 
 **Upgrade:** `pythinker update` from inside the running app — it downloads
@@ -331,26 +331,26 @@ attached to every GitHub Release.
 
 ```sh
 # Debian / Ubuntu (x86_64)
-sudo dpkg -i pythinker-code_0.58.0_amd64.deb
+sudo dpkg -i pythinker-code_0.59.0_amd64.deb
 sudo apt-get install -f       # only if dpkg reports missing deps
 
 # Debian / Ubuntu (ARM64)
-sudo dpkg -i pythinker-code_0.58.0_arm64.deb
+sudo dpkg -i pythinker-code_0.59.0_arm64.deb
 
 # Fedora / RHEL / openSUSE (x86_64)
-curl -LO https://github.com/Pythoughts-labs/pythinker-code/releases/download/v0.58.0/pythinker-code-0.58.0.x86_64.rpm
-curl -LO https://github.com/Pythoughts-labs/pythinker-code/releases/download/v0.58.0/pythinker-code-0.58.0.x86_64.rpm.sha256
-sha256sum -c pythinker-code-0.58.0.x86_64.rpm.sha256
+curl -LO https://github.com/Pythoughts-labs/pythinker-code/releases/download/v0.59.0/pythinker-code-0.59.0.x86_64.rpm
+curl -LO https://github.com/Pythoughts-labs/pythinker-code/releases/download/v0.59.0/pythinker-code-0.59.0.x86_64.rpm.sha256
+sha256sum -c pythinker-code-0.59.0.x86_64.rpm.sha256
 # Fedora / RHEL:
-sudo dnf install ./pythinker-code-0.58.0.x86_64.rpm
+sudo dnf install ./pythinker-code-0.59.0.x86_64.rpm
 # openSUSE:
-sudo zypper install ./pythinker-code-0.58.0.x86_64.rpm
+sudo zypper install ./pythinker-code-0.59.0.x86_64.rpm
 
 # Fedora / RHEL (aarch64)
-curl -LO https://github.com/Pythoughts-labs/pythinker-code/releases/download/v0.58.0/pythinker-code-0.58.0.aarch64.rpm
-curl -LO https://github.com/Pythoughts-labs/pythinker-code/releases/download/v0.58.0/pythinker-code-0.58.0.aarch64.rpm.sha256
-sha256sum -c pythinker-code-0.58.0.aarch64.rpm.sha256
-sudo dnf install ./pythinker-code-0.58.0.aarch64.rpm
+curl -LO https://github.com/Pythoughts-labs/pythinker-code/releases/download/v0.59.0/pythinker-code-0.59.0.aarch64.rpm
+curl -LO https://github.com/Pythoughts-labs/pythinker-code/releases/download/v0.59.0/pythinker-code-0.59.0.aarch64.rpm.sha256
+sha256sum -c pythinker-code-0.59.0.aarch64.rpm.sha256
+sudo dnf install ./pythinker-code-0.59.0.aarch64.rpm
 ```
 
 Both packages drop a small `/usr/bin/pythinker` launcher that execs the real
@@ -359,8 +359,8 @@ binary under `/usr/lib/pythinker/`, so your `$PATH` stays tidy.
 **Verify before install:**
 
 ```sh
-sha256sum -c pythinker-code_0.58.0_amd64.deb.sha256        # Debian/Ubuntu
-sha256sum -c pythinker-code-0.58.0.x86_64.rpm.sha256       # Fedora/RHEL
+sha256sum -c pythinker-code_0.59.0_amd64.deb.sha256        # Debian/Ubuntu
+sha256sum -c pythinker-code-0.59.0.x86_64.rpm.sha256       # Fedora/RHEL
 ```
 
 **Upgrade:** download the new `.deb`/`.rpm` from Releases and `dpkg -i` /
@@ -440,6 +440,21 @@ For hosted Pythinker models or ACP terminal auth:
 pythinker login
 ```
 
+Z.AI exposes two independent OpenAI-compatible routes. Choose the route that owns your key;
+both may coexist, and Pythinker never infers, migrates, falls back, or retries across them:
+
+```sh
+# Coding Plan subscription
+pythinker login --z-ai-coding   # reads ZAI_CODING_API_KEY when set
+
+# Standard pay-as-you-go API
+pythinker login --z-ai-api      # reads ZAI_API_KEY when set
+```
+
+These create distinct `z-ai-coding/*` and `z-ai-api/*` model identities. See the
+[provider guide](./docs/en/configuration/providers.md#managed-zai-routes) for endpoints,
+curated GLM capabilities, and route-scoped `/usage` behavior.
+
 ### 💬 Try it out
 
 ```sh

{pythinker_code-0.58.0 → pythinker_code-0.59.0}/README.md

@@ -50,13 +50,13 @@ It speaks the [**Agent Client Protocol (ACP)**](https://github.com/agentclientpr
 
 ---
 
-## 🆕 What's New in 0.58.0
+## 🆕 What's New in 0.59.0
 
-- **Agent core deepened with an observable request-assembly path.** Provider handoff, prompt ordering, and context persistence now run through explicit compatibility contracts, conversation history updates are transactional (atomic replacement with cancellation and rollback), and agent/skill definitions resolve through one source-aware catalogue instead of scattered lookups.
-- **Agent-spec loading is more defensive and truthful.** Subagent `path`, `extend`, and `system_prompt_path` references that resolve outside their spec's directory are now rejected instead of loaded, and malformed frontmatter parsing errors are no longer misreported as harmless skips.
-- **Thinking and subagent activity render cleanly in the terminal.** Live reasoning previews render complete Markdown without exposing stray HTML comments, and activity-tree rows stay visually stable during long-running turns.
+- **Explicit Z.AI compatibility routes.** Independent Z.AI Coding Plan and API login routes now use separate credentials, endpoints, model identities, catalog refresh, logout, and usage/rate-limit state, with immutable compatibility profiles keeping request-format quirks behind the chat-provider boundary.
+- **Streamed tool calls are correlated safely.** Interleaved argument chunks stay attached to their indexed calls, malformed or truncated call streams stop before tool execution, and failed attempts are not retried after output has already been shown.
+- **Deterministic reviewer scopes and supervised tool batches.** Reviewer subagents receive fully resolved Git commit anchors with invalid or empty scopes rejected explicitly, and tool execution runs through a supervised batch engine with ordered results, bounded cancellation, and fail-closed new batches until cleanup drains.
 
-Upgrade with `pythinker update`, `pip install --upgrade pythinker-code==0.58.0`, or use the native installer for your platform from the [Releases page](https://github.com/Pythoughts-labs/pythinker-code/releases/latest).
+Upgrade with `pythinker update`, `pip install --upgrade pythinker-code==0.59.0`, or use the native installer for your platform from the [Releases page](https://github.com/Pythoughts-labs/pythinker-code/releases/latest).
 
 
 ---
@@ -162,7 +162,7 @@ matches your OS — no Python, Node, or `uv` prerequisite.
 
 | Platform | Recommended install | Artifact source |
 |---|---|---|
-| **🪟 Windows** | `irm https://pythinker.com/install.ps1 \| iex` | `PythinkerSetup-0.58.0.exe` from [Releases](https://github.com/Pythoughts-labs/pythinker-code/releases/latest) |
+| **🪟 Windows** | `irm https://pythinker.com/install.ps1 \| iex` | `PythinkerSetup-0.59.0.exe` from [Releases](https://github.com/Pythoughts-labs/pythinker-code/releases/latest) |
 | **<img src="https://img.shields.io/badge/-macOS-000000?style=flat-square&logo=apple&logoColor=white" alt="macOS"> / <img src="https://img.shields.io/badge/-Linux-FCC624?style=flat-square&logo=linux&logoColor=black" alt="Linux">** | `curl -fsSL https://pythinker.com/install.sh \| bash` | native tarball from [Releases](https://github.com/Pythoughts-labs/pythinker-code/releases/latest) |
 | **<img src="https://img.shields.io/badge/-macOS-000000?style=flat-square&logo=apple&logoColor=white" alt="macOS"> — Homebrew** | `brew install Pythoughts-labs/pythinker/pythinker-code` | auto-published Homebrew tap |
 | **🐳 Docker** | `docker run --rm -it ghcr.io/pythoughts-labs/pythinker-code` | GHCR multi-arch image |
@@ -190,7 +190,7 @@ pythinker                      # start the interactive TUI
 
 ### 🪟 Windows — native installer
 
-`PythinkerSetup-0.58.0.exe` is an Inno Setup wizard. Release builds are signed
+`PythinkerSetup-0.59.0.exe` is an Inno Setup wizard. Release builds are signed
 when Authenticode secrets are configured in CI; otherwise the installer ships
 unsigned. Installs per-user into `%LOCALAPPDATA%\Programs\Pythinker`, registers
 `pythinker` on your user PATH (`HKCU\Environment`), broadcasts
@@ -202,13 +202,13 @@ irm https://pythinker.com/install.ps1 | iex
 
 # Or manually download the installer + checksum from the Releases page,
 # verify with Get-FileHash, then run:
-.\PythinkerSetup-0.58.0.exe
+.\PythinkerSetup-0.59.0.exe
 
 # Open a fresh PowerShell
 pythinker --version
 ```
 
-**Per-machine install** (IT-managed boxes): `.\PythinkerSetup-0.58.0.exe /ALLUSERS`
+**Per-machine install** (IT-managed boxes): `.\PythinkerSetup-0.59.0.exe /ALLUSERS`
 installs to `%ProgramFiles%\Pythinker` and writes PATH to HKLM (requires admin).
 
 **Upgrade:** `pythinker update` from inside the running app — it downloads
@@ -266,26 +266,26 @@ attached to every GitHub Release.
 
 ```sh
 # Debian / Ubuntu (x86_64)
-sudo dpkg -i pythinker-code_0.58.0_amd64.deb
+sudo dpkg -i pythinker-code_0.59.0_amd64.deb
 sudo apt-get install -f       # only if dpkg reports missing deps
 
 # Debian / Ubuntu (ARM64)
-sudo dpkg -i pythinker-code_0.58.0_arm64.deb
+sudo dpkg -i pythinker-code_0.59.0_arm64.deb
 
 # Fedora / RHEL / openSUSE (x86_64)
-curl -LO https://github.com/Pythoughts-labs/pythinker-code/releases/download/v0.58.0/pythinker-code-0.58.0.x86_64.rpm
-curl -LO https://github.com/Pythoughts-labs/pythinker-code/releases/download/v0.58.0/pythinker-code-0.58.0.x86_64.rpm.sha256
-sha256sum -c pythinker-code-0.58.0.x86_64.rpm.sha256
+curl -LO https://github.com/Pythoughts-labs/pythinker-code/releases/download/v0.59.0/pythinker-code-0.59.0.x86_64.rpm
+curl -LO https://github.com/Pythoughts-labs/pythinker-code/releases/download/v0.59.0/pythinker-code-0.59.0.x86_64.rpm.sha256
+sha256sum -c pythinker-code-0.59.0.x86_64.rpm.sha256
 # Fedora / RHEL:
-sudo dnf install ./pythinker-code-0.58.0.x86_64.rpm
+sudo dnf install ./pythinker-code-0.59.0.x86_64.rpm
 # openSUSE:
-sudo zypper install ./pythinker-code-0.58.0.x86_64.rpm
+sudo zypper install ./pythinker-code-0.59.0.x86_64.rpm
 
 # Fedora / RHEL (aarch64)
-curl -LO https://github.com/Pythoughts-labs/pythinker-code/releases/download/v0.58.0/pythinker-code-0.58.0.aarch64.rpm
-curl -LO https://github.com/Pythoughts-labs/pythinker-code/releases/download/v0.58.0/pythinker-code-0.58.0.aarch64.rpm.sha256
-sha256sum -c pythinker-code-0.58.0.aarch64.rpm.sha256
-sudo dnf install ./pythinker-code-0.58.0.aarch64.rpm
+curl -LO https://github.com/Pythoughts-labs/pythinker-code/releases/download/v0.59.0/pythinker-code-0.59.0.aarch64.rpm
+curl -LO https://github.com/Pythoughts-labs/pythinker-code/releases/download/v0.59.0/pythinker-code-0.59.0.aarch64.rpm.sha256
+sha256sum -c pythinker-code-0.59.0.aarch64.rpm.sha256
+sudo dnf install ./pythinker-code-0.59.0.aarch64.rpm
 ```
 
 Both packages drop a small `/usr/bin/pythinker` launcher that execs the real
@@ -294,8 +294,8 @@ binary under `/usr/lib/pythinker/`, so your `$PATH` stays tidy.
 **Verify before install:**
 
 ```sh
-sha256sum -c pythinker-code_0.58.0_amd64.deb.sha256        # Debian/Ubuntu
-sha256sum -c pythinker-code-0.58.0.x86_64.rpm.sha256       # Fedora/RHEL
+sha256sum -c pythinker-code_0.59.0_amd64.deb.sha256        # Debian/Ubuntu
+sha256sum -c pythinker-code-0.59.0.x86_64.rpm.sha256       # Fedora/RHEL
 ```
 
 **Upgrade:** download the new `.deb`/`.rpm` from Releases and `dpkg -i` /
@@ -375,6 +375,21 @@ For hosted Pythinker models or ACP terminal auth:
 pythinker login
 ```
 
+Z.AI exposes two independent OpenAI-compatible routes. Choose the route that owns your key;
+both may coexist, and Pythinker never infers, migrates, falls back, or retries across them:
+
+```sh
+# Coding Plan subscription
+pythinker login --z-ai-coding   # reads ZAI_CODING_API_KEY when set
+
+# Standard pay-as-you-go API
+pythinker login --z-ai-api      # reads ZAI_API_KEY when set
+```
+
+These create distinct `z-ai-coding/*` and `z-ai-api/*` model identities. See the
+[provider guide](./docs/en/configuration/providers.md#managed-zai-routes) for endpoints,
+curated GLM capabilities, and route-scoped `/usage` behavior.
+
 ### 💬 Try it out
 
 ```sh

{pythinker_code-0.58.0 → pythinker_code-0.59.0}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "pythinker-code"
-version = "0.58.0"
+version = "0.59.0"
 description = "Pythinker — an agentic CLI developed by Pythoughts-labs."
 readme = "README.md"
 requires-python = ">=3.12"

{pythinker_code-0.58.0 → pythinker_code-0.59.0}/src/pythinker_code/CHANGELOG.md

@@ -15,6 +15,18 @@ GitHub Releases page; `0.8.0` is the new starting line.
 
 ## Unreleased
 
+## 0.59.0 (2026-07-17)
+
+- **Reviewer subagents now receive deterministic Git scopes.** Structured automatic,
+  uncommitted, base, and commit targets resolve to full commit anchors before dispatch, reject
+  invalid or empty scopes explicitly, and keep repository metadata isolated from instructions
+  across foreground and background runs.
+- **Parallel streamed tool calls are now correlated safely.** Interleaved argument chunks stay attached to their indexed calls, malformed or truncated call streams stop before tool execution, and failed attempts are not retried after output has already been shown.
+- **Provider compatibility and Z.AI routing are now explicit.** Immutable compatibility profiles keep request-format quirks behind the chat-provider boundary, while independent Z.AI Coding Plan and API login routes use separate credentials, endpoints, model identities, catalog refresh, logout, and usage/rate-limit state. Curated GLM requests now apply exact context/output limits, thinking controls, reasoning replay, and tool-stream support without activating for local or unknown models.
+- **Tool execution is now supervised as a terminal batch.** A private execution engine preserves the Toolset registry and legacy per-call API while centralizing ordered results, deduplication, callbacks, and batch summaries; cancellation is bounded, late work stays owned, and new batches fail closed until timed-out cleanup drains.
+
+Upgrade with `pythinker update`, `pip install --upgrade pythinker-code==0.59.0`, or use the native installer for your platform from the [Releases page](https://github.com/Pythoughts-labs/pythinker-code/releases/latest).
+
 ## 0.58.0 (2026-07-11)
 
 - **Agent-spec loading is more defensive and truthful.** Subagent `path`, `extend`, and

{pythinker_code-0.58.0 → pythinker_code-0.59.0}/src/pythinker_code/auth/__init__.py

@@ -13,7 +13,8 @@ ANTHROPIC_PLATFORM_ID = "anthropic"
 OPENROUTER_PLATFORM_ID = "openrouter"
 LM_STUDIO_PLATFORM_ID = "lm-studio"
 OLLAMA_PLATFORM_ID = "ollama"
-ZAI_PLATFORM_ID = "z-ai"
+ZAI_CODING_PLATFORM_ID = "z-ai-coding"
+ZAI_API_PLATFORM_ID = "z-ai-api"
 
 __all__ = [
     "ALIBABA_PLATFORM_ID",
@@ -29,5 +30,6 @@ __all__ = [
     "OPENCODE_GO_PLATFORM_ID",
     "OPENROUTER_PLATFORM_ID",
     "PYTHINKER_CODE_PLATFORM_ID",
-    "ZAI_PLATFORM_ID",
+    "ZAI_API_PLATFORM_ID",
+    "ZAI_CODING_PLATFORM_ID",
 ]

{pythinker_code-0.58.0 → pythinker_code-0.59.0}/src/pythinker_code/auth/platforms.py

@@ -250,8 +250,9 @@ async def refresh_managed_models(config: Config) -> bool:
         refresh_opencode_go_models,
     )
     from pythinker_code.auth.z_ai import (
-        ZAI_PROVIDER_KEY,
-        ZaiModel,
+        ZAI_ROUTES,
+        ZaiCatalogResult,
+        ZaiRoute,
         apply_z_ai_models,
         refresh_z_ai_models,
     )
@@ -264,16 +265,17 @@ async def refresh_managed_models(config: Config) -> bool:
 
     changed = False
     updates: list[tuple[str, str, list[ModelInfo]]] = []
+    z_ai_provider_keys = {route.provider_key for route in ZAI_ROUTES}
     oauth_manager = None
     for provider_key, provider in managed_providers.items():
         # OpenCode Go and MiniMax own provider-specific model discovery. The
         # generic `managed:<platform>` path can't express OpenCode Go's
         # two-provider split, and MiniMax's provider key intentionally includes
         # the wire-shape suffix (`managed:minimax-anthropic`).
-        if provider_key in OPENCODE_GO_PROVIDER_KEYS or provider_key in (
-            MINIMAX_ANTHROPIC_PROVIDER_KEY,
-            ZAI_PROVIDER_KEY,
-            KIMI_PROVIDER_KEY,
+        if (
+            provider_key in OPENCODE_GO_PROVIDER_KEYS
+            or provider_key in {MINIMAX_ANTHROPIC_PROVIDER_KEY, KIMI_PROVIDER_KEY}
+            or provider_key in z_ai_provider_keys
         ):
             continue
         platform_id = parse_managed_provider_key(provider_key)
@@ -430,13 +432,29 @@ async def refresh_managed_models(config: Config) -> bool:
     if minimax_models is not None and apply_minimax_models(config, minimax_models):
         changed = True
 
-    z_ai_models: tuple[ZaiModel, ...] | None = None
-    try:
-        z_ai_models = await refresh_z_ai_models(config)
-    except (aiohttp.ClientError, TimeoutError, ValueError) as exc:
-        logger.warning("Failed to refresh Z AI models: {error}", error=exc)
-    if z_ai_models is not None and apply_z_ai_models(config, z_ai_models):
-        changed = True
+    z_ai_results: dict[ZaiRoute, ZaiCatalogResult] = {}
+    for route in ZAI_ROUTES:
+        try:
+            result = await refresh_z_ai_models(config, route)
+        except Exception as exc:
+            logger.warning(
+                "Unexpected Z.AI catalog refresh failure for {route}: {error_type}",
+                route=route.platform_id,
+                error_type=type(exc).__name__,
+            )
+            continue
+        z_ai_results[route] = result
+        if result.status == "live":
+            assert result.models is not None
+            if apply_z_ai_models(config, route, result.models):
+                changed = True
+        elif result.status != "unconfigured":
+            logger.warning(
+                "Z.AI catalog refresh did not update {route}: status={status}, failure={failure}",
+                route=route.platform_id,
+                status=result.status,
+                failure=result.failure,
+            )
 
     kimi_models: tuple[KimiModel, ...] | None = None
     try:
@@ -456,8 +474,12 @@ async def refresh_managed_models(config: Config) -> bool:
             save_changed = True
         if minimax_models is not None and apply_minimax_models(config_for_save, minimax_models):
             save_changed = True
-        if z_ai_models is not None and apply_z_ai_models(config_for_save, z_ai_models):
-            save_changed = True
+        for route, result in z_ai_results.items():
+            if result.status != "live":
+                continue
+            assert result.models is not None
+            if apply_z_ai_models(config_for_save, route, result.models):
+                save_changed = True
         if kimi_models is not None and apply_kimi_models(config_for_save, kimi_models):
             save_changed = True
         if save_changed: