Phase 5: AI provider system with local and cloud support

- Implement AIProvider base interface with chat() and is_available() - Add LocalProvider connecting to bundled llama-server via OpenAI SDK - Add OpenAIProvider for direct OpenAI API access - Add AnthropicProvider for Anthropic Claude API - Add LiteLLMProvider for multi-provider gateway - Build AIProviderService with provider routing, auto-selection, and transcript context injection - Add ai.chat IPC handler supporting chat, list_providers, set_provider, and configure actions - Add ai_chat, ai_list_providers, ai_configure Tauri commands - Build interactive AIChatPanel with message history, quick actions (Summarize, Action Items), and transcript context awareness - Tests: 30 Python, 6 Rust, 0 Svelte errors Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-26 16:25:10 -08:00
parent 415a648a2b
commit d67625cd5a
11 changed files with 735 additions and 28 deletions
--- a/python/voice_to_notes/providers/anthropic_provider.py
+++ b/python/voice_to_notes/providers/anthropic_provider.py
@@ -2,4 +2,68 @@

 from __future__ import annotations

-# TODO: Implement Anthropic provider
+import os
+from typing import Any
+
+from voice_to_notes.providers.base import AIProvider
+
+
+class AnthropicProvider(AIProvider):
+    """Connects to the Anthropic API."""
+
+    def __init__(
+        self,
+        api_key: str | None = None,
+        model: str = "claude-sonnet-4-6",
+    ) -> None:
+        self._api_key = api_key or os.environ.get("ANTHROPIC_API_KEY", "")
+        self._model = model
+        self._client: Any = None
+
+    def _ensure_client(self) -> Any:
+        if self._client is not None:
+            return self._client
+
+        if not self._api_key:
+            raise RuntimeError(
+                "Anthropic API key not configured. Set ANTHROPIC_API_KEY or provide it in settings."
+            )
+
+        try:
+            import anthropic
+
+            self._client = anthropic.Anthropic(api_key=self._api_key)
+        except ImportError:
+            raise RuntimeError("anthropic package is required. Install with: pip install anthropic")
+        return self._client
+
+    def chat(self, messages: list[dict[str, str]], **kwargs: Any) -> str:
+        client = self._ensure_client()
+
+        # Anthropic expects a system message separately
+        system_msg = ""
+        chat_messages = []
+        for msg in messages:
+            if msg.get("role") == "system":
+                system_msg = msg.get("content", "")
+            else:
+                chat_messages.append(msg)
+
+        create_kwargs: dict[str, Any] = {
+            "model": kwargs.get("model", self._model),
+            "messages": chat_messages,
+            "max_tokens": kwargs.get("max_tokens", 2048),
+        }
+        if system_msg:
+            create_kwargs["system"] = system_msg
+
+        response = client.messages.create(**create_kwargs)
+        # Anthropic returns content blocks
+        return "".join(block.text for block in response.content if hasattr(block, "text"))
+
+    def is_available(self) -> bool:
+        return bool(self._api_key)
+
+    @property
+    def name(self) -> str:
+        return "Anthropic"
--- a/python/voice_to_notes/providers/base.py
+++ b/python/voice_to_notes/providers/base.py
@@ -3,7 +3,6 @@
 from __future__ import annotations

 from abc import ABC, abstractmethod
-from collections.abc import AsyncIterator
 from typing import Any


@@ -11,13 +10,17 @@ class AIProvider(ABC):
    """Base interface for all AI providers."""

    @abstractmethod
-    async def chat(self, messages: list[dict[str, Any]], config: dict[str, Any]) -> str:
-        """Send a chat completion request and return the response."""
+    def chat(self, messages: list[dict[str, str]], **kwargs: Any) -> str:
+        """Send a chat completion request and return the full response text."""
        ...

    @abstractmethod
-    async def stream(
-        self, messages: list[dict[str, Any]], config: dict[str, Any]
-    ) -> AsyncIterator[str]:
-        """Send a streaming chat request, yielding tokens as they arrive."""
+    def is_available(self) -> bool:
+        """Check if this provider is configured and available."""
+        ...
+
+    @property
+    @abstractmethod
+    def name(self) -> str:
+        """Provider display name."""
        ...
--- a/python/voice_to_notes/providers/litellm_provider.py
+++ b/python/voice_to_notes/providers/litellm_provider.py
@@ -2,4 +2,41 @@

 from __future__ import annotations

-# TODO: Implement LiteLLM provider
+from typing import Any
+
+from voice_to_notes.providers.base import AIProvider
+
+
+class LiteLLMProvider(AIProvider):
+    """Routes through LiteLLM for access to 100+ LLM providers."""
+
+    def __init__(self, model: str = "gpt-4o-mini", **kwargs: Any) -> None:
+        self._model = model
+        self._extra_kwargs = kwargs
+
+    def chat(self, messages: list[dict[str, str]], **kwargs: Any) -> str:
+        try:
+            import litellm
+        except ImportError:
+            raise RuntimeError("litellm package is required. Install with: pip install litellm")
+
+        merged_kwargs = {**self._extra_kwargs, **kwargs}
+        response = litellm.completion(
+            model=merged_kwargs.get("model", self._model),
+            messages=messages,
+            temperature=merged_kwargs.get("temperature", 0.7),
+            max_tokens=merged_kwargs.get("max_tokens", 2048),
+        )
+        return response.choices[0].message.content or ""
+
+    def is_available(self) -> bool:
+        try:
+            import litellm  # noqa: F401
+
+            return True
+        except ImportError:
+            return False
+
+    @property
+    def name(self) -> str:
+        return "LiteLLM"
--- a/python/voice_to_notes/providers/local_provider.py
+++ b/python/voice_to_notes/providers/local_provider.py
@@ -2,8 +2,57 @@

 from __future__ import annotations

+import sys
+from typing import Any

-# TODO: Implement local provider
-# - Connect to llama-server on localhost:{port}
-# - Use openai SDK with custom base_url
-# - Support chat and streaming
+from voice_to_notes.providers.base import AIProvider
+
+
+class LocalProvider(AIProvider):
+    """Connects to bundled llama-server via its OpenAI-compatible API."""
+
+    def __init__(self, base_url: str = "http://localhost:8080", model: str = "local") -> None:
+        self._base_url = base_url.rstrip("/")
+        self._model = model
+        self._client: Any = None
+
+    def _ensure_client(self) -> Any:
+        if self._client is not None:
+            return self._client
+
+        try:
+            from openai import OpenAI
+
+            self._client = OpenAI(
+                base_url=f"{self._base_url}/v1",
+                api_key="not-needed",  # llama-server doesn't require an API key
+            )
+        except ImportError:
+            raise RuntimeError(
+                "openai package is required for local AI. Install with: pip install openai"
+            )
+        return self._client
+
+    def chat(self, messages: list[dict[str, str]], **kwargs: Any) -> str:
+        client = self._ensure_client()
+        response = client.chat.completions.create(
+            model=self._model,
+            messages=messages,
+            temperature=kwargs.get("temperature", 0.7),
+            max_tokens=kwargs.get("max_tokens", 2048),
+        )
+        return response.choices[0].message.content or ""
+
+    def is_available(self) -> bool:
+        try:
+            import urllib.request
+
+            req = urllib.request.Request(f"{self._base_url}/health", method="GET")
+            with urllib.request.urlopen(req, timeout=2) as resp:
+                return resp.status == 200
+        except Exception:
+            return False
+
+    @property
+    def name(self) -> str:
+        return "Local (llama-server)"
--- a/python/voice_to_notes/providers/openai_provider.py
+++ b/python/voice_to_notes/providers/openai_provider.py
@@ -2,4 +2,52 @@

 from __future__ import annotations

-# TODO: Implement OpenAI provider
+import os
+from typing import Any
+
+from voice_to_notes.providers.base import AIProvider
+
+
+class OpenAIProvider(AIProvider):
+    """Connects to the OpenAI API."""
+
+    def __init__(
+        self,
+        api_key: str | None = None,
+        model: str = "gpt-4o-mini",
+    ) -> None:
+        self._api_key = api_key or os.environ.get("OPENAI_API_KEY", "")
+        self._model = model
+        self._client: Any = None
+
+    def _ensure_client(self) -> Any:
+        if self._client is not None:
+            return self._client
+
+        if not self._api_key:
+            raise RuntimeError("OpenAI API key not configured. Set OPENAI_API_KEY or provide it in settings.")
+
+        try:
+            from openai import OpenAI
+
+            self._client = OpenAI(api_key=self._api_key)
+        except ImportError:
+            raise RuntimeError("openai package is required. Install with: pip install openai")
+        return self._client
+
+    def chat(self, messages: list[dict[str, str]], **kwargs: Any) -> str:
+        client = self._ensure_client()
+        response = client.chat.completions.create(
+            model=kwargs.get("model", self._model),
+            messages=messages,
+            temperature=kwargs.get("temperature", 0.7),
+            max_tokens=kwargs.get("max_tokens", 2048),
+        )
+        return response.choices[0].message.content or ""
+
+    def is_available(self) -> bool:
+        return bool(self._api_key)
+
+    @property
+    def name(self) -> str:
+        return "OpenAI"