{ "spec_version": "1.4.0-otel", "timestamp": "2026-04-24T00:13:03.355457Z", "trace_id": "4bf92f3577b34da6a3ce929d0e0e4736", "span_id": "00f067aa0ba902b7", "severity_text": "WARN", "severity_number": 13, "event_name": "swarm.agent.lifecycle.deprecated", "resource": { "service.name": "swarm-orchestrator", "service.version": "4.2.0", "deployment.environment": "production", "node.id": "worker-node-aws-us-east-1a-04" }, "attributes": { "agent.id": "RAGAgent", "agent.previous_version": "v29.x", "agent.target_version": "v30", "deprecation.reason": "Transitioning to v30 architecture.", "deprecation.action_required": true }, "metrics": { "agent.core_logic.function_count": 9, "agent.capabilities.count": 0 }, "artifacts": { "provenance": { "repository": "git+ssh://repo/core/agents/rag_agent.py", "commit_sha": "a1b2c3d4e5f6g7h8i9j0", "docstring_hash": "sha256:d8c4b12389a..." }, "extracted_capabilities": [], "core_logic_signatures": [ { "name": "__init__", "ast_hash": "sha256:cd8ca2787b78cc42", "loc": 28, "source": "def __init__(\n self,\n config: Dict[str, Any],\n constitution: Optional[Dict[str, Any]] = None,\n kernel: Optional[Kernel] = None,\n llm_engine: Optional[BaseLLMEngine] = None,\n embedding_model: Optional[BaseEmbeddingModel] = None,\n vector_store: Optional[BaseVectorStore] = None,\n **kwargs\n ):\n \"\"\"\n Initializes the RAGAgent.\n \"\"\"\n super().__init__(config, constitution, kernel)\n\n self.llm_engine = llm_engine\n self.embedding_model = embedding_model\n self.vector_store = vector_store\n self.tools = {}\n self.state = \"idle\"\n\n # Log initialization details\n log_message = f\"RAG Agent {type(self).__name__} initialized\"\n if self.kernel:\n log_message += \" with Semantic Kernel instance.\"\n else:\n log_message += \".\"\n logging.info(log_message)" }, { "name": "execute", "ast_hash": "sha256:01ed1af58d83a2d3", "loc": 20, "source": "async def execute(self, skill_name: str, **kwargs) -> Optional[Any]:\n \"\"\"\n Executes a RAG skill based on the skill_name.\n \"\"\"\n if skill_name == 'process_rag_query':\n query = kwargs.get('query')\n if not query:\n return {\"error\": \"Query not provided for process_rag_query skill.\"}\n return await self.process_query(query)\n\n elif skill_name == 'ingest_document':\n doc_input = kwargs.get('document')\n if not doc_input:\n return {\"error\": \"Document not provided for ingest_document skill.\"}\n await self.ingest_document(doc_input)\n return {\"status\": \"ingestion_complete\"}\n\n else:\n logging.warning(f\"Unknown skill '{skill_name}' requested from RAGAgent.\")\n return None" }, { "name": "process_query", "ast_hash": "sha256:0cfc9389e07ef894", "loc": 33, "source_omitted_for_brevity": true, "note": "In a real telemetry pipeline, large functions can be truncated or omitted entirely in favor of the 'ast_hash' to save network bandwidth and storage costs." }, { "name": "ingest_document", "ast_hash": "sha256:060dae0873772d80", "loc": 60, "source_omitted_for_brevity": true, "note": "In a real telemetry pipeline, large functions can be truncated or omitted entirely in favor of the 'ast_hash' to save network bandwidth and storage costs." }, { "name": "enhance_query_with_sk", "ast_hash": "sha256:fe91ea7527fa76aa", "loc": 27, "source": "async def enhance_query_with_sk(self, query: str) -> str:\n \"\"\"\n Enhances a query using a Semantic Kernel skill.\n \"\"\"\n if not self.kernel:\n logging.warning(f\"RAG Agent: Semantic Kernel not available. Returning original query.\")\n return query\n\n skill_name = \"QueryEnhancerSkill\"\n function_name = \"enhance\"\n\n try:\n # Check if plugin exists, if not try to load (mock path logic preserved from original)\n if hasattr(self.kernel, 'plugins') and skill_name not in self.kernel.plugins:\n # Logic to load would go here, omitting specific path assumptions for now\n pass\n\n if hasattr(self.kernel, 'plugins') and skill_name in self.kernel.plugins:\n enhancer_function = self.kernel.plugins[skill_name][function_name]\n # Assuming invoke takes arguments\n result = await self.kernel.invoke(enhancer_function, query=query)\n return str(result)\n\n return query\n except Exception as e:\n logging.error(f\"RAG Agent: Error running QueryEnhancerSkill: {e}\")\n return query" }, { "name": "register_tool", "ast_hash": "sha256:ea3c1b0820bf857f", "loc": 22, "source": "def register_tool(self, tool_instance: Any, plugin_name: Optional[str] = None):\n \"\"\"\n Registers a tool with the agent's Semantic Kernel instance.\n \"\"\"\n if not self.kernel:\n logging.warning(f\"RAG Agent: Cannot register tool, Semantic Kernel not available.\")\n return\n\n if not hasattr(tool_instance, \"name\"):\n logging.error(f\"RAG Agent: Tool instance does not have a 'name' attribute.\")\n return\n\n tool_name = plugin_name or tool_instance.name\n\n try:\n # SK v1.x\n if hasattr(self.kernel, 'add_plugin'):\n self.kernel.add_plugin(plugin_instance=tool_instance, plugin_name=tool_name)\n self.tools[tool_name] = tool_instance\n logging.info(f\"RAG Agent: Registered tool '{tool_name}' with Semantic Kernel.\")\n except Exception as e:\n logging.error(f\"RAG Agent: Failed to register tool '{tool_name}': {e}\")" }, { "name": "invoke_tool", "ast_hash": "sha256:25cffbc983a78701", "loc": 16, "source": "async def invoke_tool(self, plugin_name: str, function_name: str, **kwargs) -> Optional[str]:\n \"\"\"\n Invokes a registered tool's function via Semantic Kernel.\n \"\"\"\n if not self.kernel:\n logging.warning(f\"RAG Agent: Cannot invoke tool, Semantic Kernel not available.\")\n return None\n\n try:\n if hasattr(self.kernel, 'plugins') and plugin_name in self.kernel.plugins:\n target_function = self.kernel.plugins[plugin_name][function_name]\n result = await self.kernel.invoke(target_function, **kwargs)\n return str(result)\n except Exception as e:\n logging.error(f\"RAG Agent: Error invoking tool '{plugin_name}.{function_name}': {e}\")\n return None" }, { "name": "search_web_if_needed", "ast_hash": "sha256:4f1ee6b321bcd626", "loc": 16, "source": "async def search_web_if_needed(self, query: str, direct_url: Optional[str] = None) -> Optional[str]:\n \"\"\"\n Example of how an agent might decide to use the web search tool.\n \"\"\"\n if \"search for\" in query.lower() or \"find on the web\" in query.lower() or direct_url:\n logging.info(f\"RAG Agent: Web search triggered for query: '{query}'\")\n plugin_name = \"web_search\"\n function_name = \"fetch_web_content\"\n tool_args = {}\n if direct_url:\n tool_args[\"url\"] = direct_url\n if query and not direct_url:\n tool_args[\"query\"] = query\n\n return await self.invoke_tool(plugin_name, function_name, **tool_args)\n return None" }, { "name": "get_skill_schema", "ast_hash": "sha256:02b07d2006b3a083", "loc": 24, "source": "def get_skill_schema(self) -> Dict[str, Any]:\n \"\"\"\n Defines the agent's skills (MCP).\n \"\"\"\n return {\n \"name\": \"RAGAgent\",\n \"description\": \"Handles document ingestion and retrieval-augmented generation queries.\",\n \"skills\": [\n {\n \"name\": \"process_rag_query\",\n \"description\": \"Answers a query based on documents in the vector store.\",\n \"parameters\": [\n {\"name\": \"query\", \"type\": \"string\", \"description\": \"The user's query.\"}\n ]\n },\n {\n \"name\": \"ingest_document\",\n \"description\": \"Ingests a document into the RAG system's vector store.\",\n \"parameters\": [\n {\"name\": \"document\", \"type\": \"any\", \"description\": \"The document content or object to ingest.\"}\n ]\n }\n ]\n }" } ] } }