feat(agent-core): 增加智能编排与模型工具基础

agent_core/orchestrator.py

@@ -0,0 +1,40 @@
+import time
+
+from .results import AgentResult
+from .structured_output import build_mock_structured_output
+from .tool_registry import run_declared_tools
+from .rag.retriever import retrieve
+
+
+def run_agent(scenario_config: dict, user_input: str, options: dict | None = None) -> AgentResult:
+    started_at = time.perf_counter()
+    options = options or {}
+    output_type = scenario_config.get("output", {}).get("type", "general_answer")
+
+    references = []
+    rag_config = scenario_config.get("rag", {})
+    if rag_config.get("enabled"):
+        references = retrieve(
+            scenario_id=scenario_config.get("id", ""),
+            query=user_input,
+            collection=rag_config.get("collection", scenario_config.get("id", "")),
+            top_k=rag_config.get("top_k", 5),
+            document_ids=options.get("document_ids"),
+            store_path=options.get("rag_store_path"),
+        )
+
+    tool_calls = run_declared_tools(scenario_config.get("tools", []), user_input)
+    structured_output = build_mock_structured_output(output_type, user_input, references)
+    answer = f"已根据「{scenario_config.get('name', '当前场景')}」生成模拟回答：{user_input}"
+    latency_ms = int((time.perf_counter() - started_at) * 1000)
+
+    return AgentResult(
+        answer=answer,
+        structured_output=structured_output,
+        references=references,
+        tool_calls=tool_calls,
+        raw_output=answer,
+        model_name=options.get("model_name", "mock-model"),
+        latency_ms=latency_ms,
+        status="success",
+    )