coderouter-cli 2.4.0__py3-none-any.whl → 2.5.1__py3-none-any.whl

coderouter/cli.py

@@ -61,7 +61,7 @@ def _build_parser() -> argparse.ArgumentParser:
             "binding the server. Repeat to layer multiple files. By "
             "default, file values do NOT override variables already in "
             "the environment (the shell `export` wins). See "
-            "docs/troubleshooting.md §5 for 1Password / direnv / sops "
+            "docs/guides/troubleshooting.md §5 for 1Password / direnv / sops "
             "integration recipes."
         ),
     )
@@ -116,7 +116,7 @@ def _build_parser() -> argparse.ArgumentParser:
             "POSIX file mode (0600 expected), .gitignore coverage, "
             "and git-tracking state. Bare `--check-env` (no PATH) "
             "looks for `./.env` then `~/.coderouter/.env`. "
-            "See docs/troubleshooting.md §5 for the threat model."
+            "See docs/guides/troubleshooting.md §5 for the threat model."
         ),
     )
     doctor.add_argument(
@@ -699,8 +699,8 @@ def _run_replay(args: argparse.Namespace) -> int:
     if getattr(args, "suggest_rules", False):
         # P1-6: statistical rule suggestion mode.
         # Always compute a full window summary (ignores --compare / --provider).
-        from coderouter.state.suggest_rules import format_suggestions, suggest_rules
         from coderouter.state.replay import summarize_window as _sw
+        from coderouter.state.suggest_rules import format_suggestions, suggest_rules
 
         # Re-read without provider filter so we see all providers.
         all_entries = read_request_log(log_path, since=args.since)

coderouter/config/capability_registry.py

@@ -109,7 +109,7 @@ class RegistryCapabilities(BaseModel):
             "harness. ``degraded`` = the model over-eagerly invokes "
             "tools/skills when given Claude Code's system prompt — e.g. "
             "Llama-3.3-70B treating small talk like ``こんにちは`` as "
-            "``Skill(hello)`` invocations (see docs/troubleshooting.md "
+            "``Skill(hello)`` invocations (see docs/guides/troubleshooting.md "
             "§4-1 for the symptom log). ``ok`` = explicitly verified "
             "clean. ``None`` = no opinion (treated as ``ok`` at the "
             "startup check)."

coderouter/config/schemas.py

@@ -874,6 +874,130 @@ class AutoRouterConfig(BaseModel):
     )
 
 
+class LauncherBackendConfig(BaseModel):
+    """Per-backend binary path configuration for the Launcher.
+
+    When ``binary`` is unset, the Launcher falls back to the default
+    executable name (``llama-server`` for llama.cpp, ``python`` for vllm)
+    and relies on ``$PATH`` resolution — which works when the tool is
+    globally installed.  Set ``binary`` when:
+
+    - llama.cpp was built from source (e.g. ``~/llama.cpp/build/bin/llama-server``)
+    - vllm lives in a virtualenv (e.g. ``~/.venv/bin/python``)
+    - Multiple builds coexist and you want to pin a specific one
+
+    Tilde (``~``) and environment variables are expanded at launch time.
+
+    Example::
+
+        backends:
+          llama.cpp:
+            binary: ~/llama.cpp/build/bin/llama-server
+          vllm:
+            binary: ~/.venv/bin/python
+    """
+
+    model_config = ConfigDict(extra="forbid")
+
+    binary: str | None = Field(
+        default=None,
+        description=(
+            "Absolute or ``~``-relative path to the backend executable. "
+            "llama.cpp default: ``llama-server`` (PATH). "
+            "vllm default: ``python`` (PATH). "
+            "Expanded at launch time."
+        ),
+    )
+
+
+class LauncherOptionProfile(BaseModel):
+    """One named option preset for a launcher backend (e.g. llama.cpp / vllm).
+
+    ``args`` maps CLI flag strings to their values.  A bool value of
+    ``True`` means "include the flag without a value" (e.g. ``--no-mmap``);
+    ``False`` means "omit the flag entirely".  All other value types are
+    converted to strings and appended as ``--flag value`` pairs.
+
+    Example::
+
+        name: "GPU速度重視"
+        args:
+          "-ngl": 99
+          "--ctx-size": 4096
+          "--no-mmap": false
+    """
+
+    model_config = ConfigDict(extra="forbid")
+
+    name: str = Field(..., description="Display name shown in the Launcher UI dropdown.")
+    args: dict[str, str | int | float | bool] = Field(
+        default_factory=dict,
+        description=(
+            "CLI flag → value mapping. "
+            "bool True = flag only (no value). "
+            "bool False = omit flag. "
+            "All other types are stringified and passed as '--flag value'."
+        ),
+    )
+
+
+class LauncherConfig(BaseModel):
+    """The ``launcher:`` block in providers.yaml.
+
+    Controls the Launcher UI available at ``/launcher``.
+
+    Example::
+
+        launcher:
+          model_dirs:
+            - ~/models
+            - /data/gguf
+          option_profiles:
+            llama.cpp:
+              - name: "GPU速度重視"
+                args:
+                  "-ngl": 99
+                  "--ctx-size": 4096
+            vllm:
+              - name: "標準"
+                args:
+                  "--dtype": "auto"
+                  "--max-model-len": 4096
+    """
+
+    model_config = ConfigDict(extra="forbid")
+
+    model_dirs: list[str] = Field(
+        default_factory=list,
+        description=(
+            "Directories to scan for model files "
+            "(.gguf, .safetensors, .bin, .pt, .ggml). "
+            "Paths are expanded (~ and env vars) at scan time, not at load. "
+            "Non-existent paths are silently skipped."
+        ),
+    )
+    backends: dict[str, LauncherBackendConfig] = Field(
+        default_factory=dict,
+        description=(
+            "Per-backend binary path overrides. "
+            "Keys are backend names ('llama.cpp', 'vllm'). "
+            "When a key is absent, the default executable is used "
+            "('llama-server' / 'python') and resolved via PATH. "
+            "Useful when running a from-source build or a venv-specific binary."
+        ),
+    )
+    option_profiles: dict[str, list[LauncherOptionProfile]] = Field(
+        default_factory=dict,
+        description=(
+            "Named option presets per backend. "
+            "Keys should be backend names: 'llama.cpp', 'vllm'. "
+            "Each key maps to an ordered list of named presets. "
+            "A free-form 'extra args' field is always available in the UI "
+            "for one-off overrides without touching this config."
+        ),
+    )
+
+
 class PluginsConfig(BaseModel):
     """The ``plugins:`` block in providers.yaml (v2.3.0).
 
@@ -1082,6 +1206,17 @@ class CodeRouterConfig(BaseModel):
             "plugins (zero-cost, backward-compatible default)."
         ),
     )
+    launcher: LauncherConfig | None = Field(
+        default=None,
+        description=(
+            "Launcher configuration for the /launcher UI. "
+            "Defines model_dirs to scan and option_profiles per backend "
+            "('llama.cpp', 'vllm'). "
+            "Unset (None) = Launcher UI shows empty model list and no profiles. "
+            "The Launcher UI itself is always available at /launcher "
+            "regardless of this setting."
+        ),
+    )
 
     @model_validator(mode="after")
     def _check_default_profile_exists(self) -> CodeRouterConfig:

coderouter/data/model-capabilities.yaml

@@ -35,7 +35,7 @@
 #                                       Claude Code's agentic-coding harness;
 #                                       "degraded" triggers a startup WARN when
 #                                       the provider is on a `claude-code-*`
-#                                       chain. See docs/troubleshooting.md §4-1.
+#                                       chain. See docs/guides/troubleshooting.md §4-1.
 #
 # First-match semantics: rules within a file are evaluated top-to-bottom
 # per flag; the first rule whose glob matches AND declares that flag
@@ -153,7 +153,7 @@ rules:
   # Llama-3.3-70B (verified 2026-04-24 against NVIDIA NIM) rewrites
   # ``こんにちは`` into ``Skill(hello)`` invocations and fabricates
   # ``AskUserQuestion("What is your name?")`` elicitations — see
-  # docs/articles/note-nvidia-nim.md §6-2 + docs/troubleshooting.md §4-1.
+  # docs/articles/note/note-nvidia-nim.md §6-2 + docs/guides/troubleshooting.md §4-1.
   #
   # Glob coverage: NIM uses ``meta/llama-3.3-70b-instruct``, OpenRouter
   # uses ``meta-llama/llama-3.3-70b-instruct``, some local servers use

coderouter/ingress/app.py

@@ -13,6 +13,7 @@ from coderouter import __version__
 from coderouter.config import load_config
 from coderouter.ingress.anthropic_routes import router as anthropic_router
 from coderouter.ingress.dashboard_routes import router as dashboard_router
+from coderouter.ingress.launcher_routes import router as launcher_router
 from coderouter.ingress.metrics_routes import router as metrics_router
 from coderouter.ingress.openai_routes import router as openai_router
 from coderouter.logging import configure_logging, get_logger
@@ -178,6 +179,12 @@ def create_app(config_path: str | None = None) -> FastAPI:
             with contextlib.suppress(Exception):
                 await probe_task
 
+        # Launcher: stop child llama.cpp / vllm processes so they don't orphan.
+        from coderouter.ingress.launcher_routes import shutdown_launcher
+
+        with contextlib.suppress(Exception):
+            await shutdown_launcher(app)
+
         # v2.0-J: graceful shutdown of recovery probe tasks.
         with contextlib.suppress(Exception):
             await engine.shutdown_recovery_probes()
@@ -259,6 +266,10 @@ def create_app(config_path: str | None = None) -> FastAPI:
     # Same root-level mount as /metrics.json — the dashboard is a UI
     # concern and doesn't belong under the /v1 API surface.
     app.include_router(dashboard_router, tags=["dashboard"])
+    # Launcher UI + process management API.
+    # /launcher       → single-page HTML UI
+    # /api/launcher/* → model scan, process start/stop/logs
+    app.include_router(launcher_router, tags=["launcher"])
 
     return app
 

coderouter/ingress/dashboard_routes.py

@@ -76,6 +76,7 @@ _DASHBOARD_HTML = r"""<!doctype html>
   <header class="border-b border-slate-800 px-6 py-3">
     <div class="max-w-7xl mx-auto flex flex-wrap items-center gap-x-6 gap-y-2 text-sm">
       <span class="text-lg font-semibold tracking-tight">CodeRouter</span>
+      <a href="/launcher" class="text-slate-400 hover:text-slate-200 transition-colors text-sm">Launcher</a>
       <span class="text-slate-400">profile: <span data-bind="profile" class="text-slate-100 font-mono">—</span></span>
       <span class="text-slate-400">uptime: <span data-bind="uptime" class="text-slate-100 font-mono tabnum">—</span></span>
       <span class="text-slate-400">requests: <span data-bind="requests_total" class="text-slate-100 font-mono tabnum">0</span></span>