Introduce Expert Agent. Prompt agents from UI.

backend/src/agents/expert.py

@@ -0,0 +1,53 @@
+import re
+import json
+import inspect
+from string import Template
+
+from inference import LlamaMessage, replicate_llama_completion, Models
+from agent import Agent, AgentField
+
+CLASS_DESCRIPTION = """
+Expert AI in any field, powered by Llama 3 70b.
+Will get guidelines, documents, and other advanced features in the future.
+""".strip()
+
+class ExpertAgent(Agent):
+    identifier = "expert"
+    description = CLASS_DESCRIPTION
+    schema_fields = [
+        AgentField('system_prompt', 'string', default="You are an expert AI assistant."),
+    ]
+
+    def __init__(self, app, data):
+        Agent.__init__(self, app, data)
+
+    async def prompt(self, instruction, input=None, **kwargs):
+        # if instruction and input are passed, instruction extends the system prompt
+        # otherwise, use the one passed as the user prompt, with the regular system prompt
+
+        # TODO: if 'history' in kwargs: interpret it as user-agent turns assuming list[str]
+
+        system = self.config['system_prompt']
+        if instruction != None and input != None:
+            system += "\n" + instruction
+
+        prompt = input if input != None else instruction
+        messages = [
+            LlamaMessage("system", system),
+            LlamaMessage("user", prompt)
+        ]
+
+        response = await replicate_llama_completion(self.app, messages, model=Models.llama_70b)
+        return response
+
+    async def received_message(self, message):
+        raise Exception("TODO: expert chat support")
+
+        messages = []
+        response = await replicate_llama_completion(self.app, messages, model=Models.llama_70b)
+        self.app.send_chat_message(
+            chat_id=message.chat_id,
+            sender=self.id,
+            content=response,
+            thread_id=message.thread_id,
+        )