solana-agent 15.1.0__tar.gz → 15.1.2__tar.gz

{solana_agent-15.1.0 → solana_agent-15.1.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: solana-agent
-Version: 15.1.0
+Version: 15.1.2
 Summary: The Future of Work
 License: MIT
 Keywords: ai,openai,ai agents,agi

{solana_agent-15.1.0 → solana_agent-15.1.2}/pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "solana-agent"
-version = "15.1.0"
+version = "15.1.2"
 description = "The Future of Work"
 authors = ["Bevan Hunt <bevan@bevanhunt.com>"]
 license = "MIT"

{solana_agent-15.1.0 → solana_agent-15.1.2}/solana_agent/adapters/llm_adapter.py

@@ -30,6 +30,8 @@ class OpenAIAdapter(LLMProvider):
         instructions: str = "",
         voice: Literal["alloy", "ash", "ballad", "coral", "echo",
                        "fable", "onyx", "nova", "sage", "shimmer"] = "nova",
+        response_format: Literal['mp3', 'opus',
+                                 'aac', 'flac', 'wav', 'pcm'] = "aac",
     ) -> AsyncGenerator[bytes, None]:  # pragma: no cover
         """Stream text-to-speech audio from OpenAI models.
 
@@ -37,6 +39,7 @@ class OpenAIAdapter(LLMProvider):
             text: Text to convert to speech
             instructions: Optional instructions for speech generation
             voice: Voice to use for synthesis
+            response_format: Audio format
 
         Yields:
             Audio bytes as they become available
@@ -47,6 +50,7 @@ class OpenAIAdapter(LLMProvider):
                 voice=voice,
                 input=text,
                 instructions=instructions,
+                response_format=response_format
             )
 
             # Stream the bytes in chunks

{solana_agent-15.1.0 → solana_agent-15.1.2}/solana_agent/client/solana_agent.py

@@ -48,6 +48,8 @@ class SolanaAgent(SolanaAgentInterface):
         voice: Literal["alloy", "ash", "ballad", "coral", "echo",
                        "fable", "onyx", "nova", "sage", "shimmer"] = "nova",
         audio_instructions: Optional[str] = None,
+        response_format: Literal['mp3', 'opus',
+                                 'aac', 'flac', 'wav', 'pcm'] = "aac",
     ) -> AsyncGenerator[Union[str, bytes], None]:  # pragma: no cover
         """Process a user message and return the response stream.
 
@@ -57,6 +59,7 @@ class SolanaAgent(SolanaAgentInterface):
             output_format: Response format ("text" or "audio")
             voice: Voice to use for audio output (only used if output_format is "audio")
             audio_instructions: Optional instructions for audio synthesis
+            response_format: Audio format
 
         Returns:
             Async generator yielding response chunks (text strings or audio bytes)

{solana_agent-15.1.0 → solana_agent-15.1.2}/solana_agent/factories/agent_factory.py

@@ -104,7 +104,7 @@ class SolanaAgentFactory:
             agent_repo.delete_ai_agent(agent_name)
 
         # Register predefined agents
-        for agent_config in config.get("ai_agents", []):
+        for agent_config in config.get("agents", []):
             agent_service.register_ai_agent(
                 name=agent_config["name"],
                 instructions=agent_config["instructions"],

{solana_agent-15.1.0 → solana_agent-15.1.2}/solana_agent/interfaces/client/client.py

@@ -14,7 +14,9 @@ class SolanaAgent(ABC):
         output_format: Literal["text", "audio"] = "text",
         voice: Literal["alloy", "ash", "ballad", "coral", "echo",
                        "fable", "onyx", "nova", "sage", "shimmer"] = "nova",
-        audio_instructions: str = None
+        audio_instructions: str = None,
+        response_format: Literal['mp3', 'opus',
+                                 'aac', 'flac', 'wav', 'pcm'] = "aac",
     ) -> AsyncGenerator[Union[str, bytes], None]:
         """Process a user message and return the response stream."""
         pass

{solana_agent-15.1.0 → solana_agent-15.1.2}/solana_agent/interfaces/providers/llm.py

@@ -39,6 +39,8 @@ class LLMProvider(ABC):
         instructions: str = "",
         voice: Literal["alloy", "ash", "ballad", "coral", "echo",
                        "fable", "onyx", "nova", "sage", "shimmer"] = "nova",
+        response_format: Literal['mp3', 'opus',
+                                 'aac', 'flac', 'wav', 'pcm'] = "aac",
     ) -> AsyncGenerator[bytes, None]:
         """Stream text-to-speech audio from the language model."""
         pass

{solana_agent-15.1.0 → solana_agent-15.1.2}/solana_agent/interfaces/services/agent.py

@@ -34,6 +34,8 @@ class AgentService(ABC):
         voice: Literal["alloy", "ash", "ballad", "coral", "echo",
                        "fable", "onyx", "nova", "sage", "shimmer"] = "nova",
         audio_instructions: str = None,
+        response_format: Literal['mp3', 'opus',
+                                 'aac', 'flac', 'wav', 'pcm'] = "aac",
     ) -> AsyncGenerator[Union[str, bytes], None]:
         """Generate a response from an agent."""
         pass

{solana_agent-15.1.0 → solana_agent-15.1.2}/solana_agent/interfaces/services/query.py

@@ -15,6 +15,8 @@ class QueryService(ABC):
         voice: Literal["alloy", "ash", "ballad", "coral", "echo",
                        "fable", "onyx", "nova", "sage", "shimmer"] = "nova",
         audio_instructions: Optional[str] = None,
+        response_format: Literal['mp3', 'opus',
+                                 'aac', 'flac', 'wav', 'pcm'] = "aac",
     ) -> AsyncGenerator[Union[str, bytes], None]:
         """Process the user request and generate a response."""
         pass

{solana_agent-15.1.0 → solana_agent-15.1.2}/solana_agent/services/agent.py

@@ -195,6 +195,8 @@ class AgentService(AgentServiceInterface):
         voice: Literal["alloy", "ash", "ballad", "coral", "echo",
                        "fable", "onyx", "nova", "sage", "shimmer"] = "nova",
         audio_instructions: Optional[str] = None,
+        response_format: Literal['mp3', 'opus',
+                                 'aac', 'flac', 'wav', 'pcm'] = "aac",
     ) -> AsyncGenerator[Union[str, bytes], None]:  # pragma: no cover
         """Generate a response with support for text/audio input/output.
 
@@ -206,6 +208,7 @@ class AgentService(AgentServiceInterface):
             output_format: Response format ("text" or "audio")
             voice: Voice to use for audio output
             audio_instructions: Optional instructions for audio synthesis
+            response_format: Audio format
 
         Yields:
             Text chunks or audio bytes depending on output_format
@@ -214,7 +217,7 @@ class AgentService(AgentServiceInterface):
         if not agent:
             error_msg = f"Agent '{agent_name}' not found."
             if output_format == "audio":
-                async for chunk in self.llm_provider.tts(error_msg, voice=voice):
+                async for chunk in self.llm_provider.tts(error_msg, instructions=audio_instructions, response_format=response_format, voice=voice):
                     yield chunk
             else:
                 yield error_msg
@@ -252,10 +255,10 @@ class AgentService(AgentServiceInterface):
                 if chunk.strip().startswith("{"):
                     # Handle tool calls
                     result = await self._handle_tool_call(
-                        agent_name, chunk, output_format, voice
+                        agent_name, chunk,
                     )
                     if output_format == "audio":
-                        async for audio_chunk in self.llm_provider.tts(result, instructions=audio_instructions, voice=voice):
+                        async for audio_chunk in self.llm_provider.tts(result, instructions=audio_instructions, response_format=response_format, voice=voice):
                             yield audio_chunk
                     else:
                         yield result
@@ -265,7 +268,7 @@ class AgentService(AgentServiceInterface):
                         text_buffer += chunk
                         if any(punct in chunk for punct in ".!?"):
                             async for audio_chunk in self.llm_provider.tts(
-                                text_buffer, instructions=audio_instructions, voice=voice
+                                text_buffer, instructions=audio_instructions, response_format=response_format, voice=voice
                             ):
                                 yield audio_chunk
                             text_buffer = ""
@@ -275,14 +278,14 @@ class AgentService(AgentServiceInterface):
             # Handle any remaining text in buffer
             if output_format == "audio" and text_buffer:
                 async for audio_chunk in self.llm_provider.tts(
-                    text_buffer, instructions=audio_instructions, voice=voice
+                    text_buffer, instructions=audio_instructions, response_format=response_format, voice=voice
                 ):
                     yield audio_chunk
 
         except Exception as e:
             error_msg = f"I apologize, but I encountered an error: {str(e)}"
             if output_format == "audio":
-                async for chunk in self.llm_provider.tts(error_msg, instructions=audio_instructions, voice=voice):
+                async for chunk in self.llm_provider.tts(error_msg, instructions=audio_instructions, response_format=response_format, voice=voice):
                     yield chunk
             else:
                 yield error_msg

{solana_agent-15.1.0 → solana_agent-15.1.2}/solana_agent/services/query.py

@@ -42,6 +42,8 @@ class QueryService(QueryServiceInterface):
         voice: Literal["alloy", "ash", "ballad", "coral", "echo",
                        "fable", "onyx", "nova", "sage", "shimmer"] = "nova",
         audio_instructions: Optional[str] = None,
+        response_format: Literal['mp3', 'opus',
+                                 'aac', 'flac', 'wav', 'pcm'] = "aac",
     ) -> AsyncGenerator[Union[str, bytes], None]:  # pragma: no cover
         """Process the user request with appropriate agent.
 
@@ -51,6 +53,7 @@ class QueryService(QueryServiceInterface):
             output_format: Response format ("text" or "audio")
             voice: Voice to use for audio output
             audio_instructions: Optional instructions for audio synthesis
+            response_format: Audio response format
 
         Yields:
             Response chunks (text strings or audio bytes)
@@ -68,7 +71,7 @@ class QueryService(QueryServiceInterface):
             if user_text.strip().lower() in ["test", "hello", "hi", "hey", "ping"]:
                 response = "Hello! How can I help you today?"
                 if output_format == "audio":
-                    async for chunk in self.agent_service.llm_provider.tts(response, instructions=audio_instructions, voice=voice):
+                    async for chunk in self.agent_service.llm_provider.tts(response, instructions=audio_instructions, response_format=response_format, voice=voice):
                         yield chunk
                 else:
                     yield response
@@ -94,7 +97,8 @@ class QueryService(QueryServiceInterface):
                 query=user_text,
                 memory_context=memory_context,
                 output_format=output_format,
-                voice=voice
+                voice=voice,
+                response_format=response_format,
             ):
                 yield chunk
                 if output_format == "text":
@@ -119,7 +123,7 @@ class QueryService(QueryServiceInterface):
         except Exception as e:
             error_msg = f"I apologize for the technical difficulty. {str(e)}"
             if output_format == "audio":
-                async for chunk in self.agent_service.llm_provider.tts(error_msg, instructions=audio_instructions, voice=voice):
+                async for chunk in self.agent_service.llm_provider.tts(error_msg, instructions=audio_instructions, response_format=response_format, voice=voice):
                     yield chunk
             else:
                 yield error_msg