update langchain examples

examples/langchain/00-math_no_approvals.py

@@ -3,40 +3,53 @@
 # with no approvals required
 #
 from dotenv import load_dotenv
+from langchain_core.prompts import ChatPromptTemplate
 from langchain_openai import ChatOpenAI
 
-from langchain.agents import AgentType, initialize_agent
-from langchain.tools import tool
+import langchain.tools
+from langchain.agents import AgentExecutor, create_tool_calling_agent
 
 load_dotenv()
 
 
-@tool
 def add(x: int, y: int) -> int:
     """Add two numbers together."""
     return x + y
 
 
-@tool
 def multiply(x: int, y: int) -> int:
     """multiply two numbers"""
     return x * y
 
 
-tools = [add.as_tool(), multiply.as_tool()]
+tools = [
+    langchain.tools.StructuredTool.from_function(add),
+    langchain.tools.StructuredTool.from_function(multiply),
+]
 
 llm = ChatOpenAI(model="gpt-4o", temperature=0)
-agent = initialize_agent(
-    tools,
-    llm,
-    agent=AgentType.OPENAI_FUNCTIONS,
-    verbose=True,
-    handle_parsing_errors=True,
+
+
+# Prompt for creating Tool Calling Agent
+prompt = ChatPromptTemplate.from_messages(
+    [
+        (
+            "system",
+            "You are a helpful assistant.",
+        ),
+        ("placeholder", "{chat_history}"),
+        ("human", "{input}"),
+        ("placeholder", "{agent_scratchpad}"),
+    ]
 )
 
+# Construct the Tool Calling Agent
+agent = create_tool_calling_agent(llm, tools, prompt)
+agent_executor = AgentExecutor(agent=agent, tools=tools, verbose=True)
 
-def main():
-    result = agent.run("multiply 2 and 5, then add 32 to the result")
+
+def main() -> None:
+    result = agent_executor.invoke({"input": "multiply 2 and 5, then add 32 to the result"})
     print("\n\n----------Result----------\n\n")
     print(result)
 

examples/langchain/00a-math_example_cli.py

@@ -5,11 +5,12 @@
 # use local / CLI mode
 #
 from dotenv import load_dotenv
+from langchain_core.prompts import ChatPromptTemplate
 from langchain_openai import ChatOpenAI
 
+import langchain.tools
 from humanlayer import HumanLayer
-from langchain.agents import AgentType, initialize_agent
-from langchain.tools import tool
+from langchain.agents import AgentExecutor, create_tool_calling_agent
 
 load_dotenv()
 
@@ -20,33 +21,43 @@ hl = HumanLayer(
 )
 
 
-@tool
 def add(x: int, y: int) -> int:
     """Add two numbers together."""
     return x + y
 
 
-@tool
 @hl.require_approval()
 def multiply(x: int, y: int) -> int:
     """multiply two numbers"""
     return x * y
 
 
-tools = [add.as_tool(), multiply.as_tool()]
+tools = [
+    langchain.tools.StructuredTool.from_function(add),
+    langchain.tools.StructuredTool.from_function(multiply),
+]
 
-llm = ChatOpenAI(model="gpt-4o", temperature=0)
-agent = initialize_agent(
-    tools,
-    llm,
-    agent=AgentType.OPENAI_FUNCTIONS,
-    verbose=True,
-    handle_parsing_errors=True,
+llm = ChatOpenAI(model="gpt-4", temperature=0)
+
+# Prompt for creating Tool Calling Agent
+prompt = ChatPromptTemplate.from_messages(
+    [
+        (
+            "system",
+            "You are a helpful assistant.",
+        ),
+        ("placeholder", "{chat_history}"),
+        ("human", "{input}"),
+        ("placeholder", "{agent_scratchpad}"),
+    ]
 )
 
+# Construct the Tool Calling Agent
+agent = create_tool_calling_agent(llm, tools, prompt)
+agent_executor = AgentExecutor(agent=agent, tools=tools, verbose=True)
 
-def main():
-    result = agent.run("multiply 2 and 5, then add 32 to the result")
+def main() -> None:
+    result = agent_executor.invoke({"input": "multiply 2 and 5, then add 32 to the result"})
     print("\n\n----------Result----------\n\n")
     print(result)
 

examples/langchain/01-math_example.py

@@ -5,10 +5,11 @@
 #
 from dotenv import load_dotenv
 from langchain_openai import ChatOpenAI
+from langchain.agents import AgentExecutor, create_tool_calling_agent
+from langchain_core.prompts import ChatPromptTemplate
+import langchain.tools
 
 from humanlayer.core.approval import HumanLayer
-from langchain.agents import AgentType, initialize_agent
-from langchain.tools import tool
 
 load_dotenv()
 
@@ -19,33 +20,45 @@ hl = HumanLayer(
 )
 
 
-@tool
 def add(x: int, y: int) -> int:
     """Add two numbers together."""
     return x + y
 
 
-@tool
 @hl.require_approval()
 def multiply(x: int, y: int) -> int:
     """multiply two numbers"""
     return x * y
 
 
-tools = [add.as_tool(), multiply.as_tool()]
+tools = [
+    langchain.tools.StructuredTool.from_function(add),
+    langchain.tools.StructuredTool.from_function(multiply),
+]
 
 llm = ChatOpenAI(model="gpt-4o", temperature=0)
-agent = initialize_agent(
-    tools,
-    llm,
-    agent=AgentType.OPENAI_FUNCTIONS,
-    verbose=True,
-    handle_parsing_errors=True,
+
+
+# Prompt for creating Tool Calling Agent
+prompt = ChatPromptTemplate.from_messages(
+    [
+        (
+            "system",
+            "You are a helpful assistant.",
+        ),
+        ("placeholder", "{chat_history}"),
+        ("human", "{input}"),
+        ("placeholder", "{agent_scratchpad}"),
+    ]
 )
 
+# Construct the Tool Calling Agent
+agent = create_tool_calling_agent(llm, tools, prompt)
+agent_executor = AgentExecutor(agent=agent, tools=tools, verbose=True)
 
-def main():
-    result = agent.run("multiply 2 and 5, then add 32 to the result")
+
+def main() -> None:
+    result = agent_executor.invoke({"input": "multiply 2 and 5, then add 32 to the result"})
     print("\n\n----------Result----------\n\n")
     print(result)
 

examples/ts_email_classifier/package-lock.json

@@ -22,21 +22,6 @@
         "typescript": "^5.5.2"
       }
     },
-    "../../humanlayer-ts": {
-      "name": "humanlayer",
-      "version": "0.7.1",
-      "license": "Apache-2.0",
-      "devDependencies": {
-        "@types/jest": "^29.5.14",
-        "@types/node": "^20.17.11",
-        "eslint": "^8.57.0",
-        "jest": "^29.7.0",
-        "prettier": "^3.3.2",
-        "ts-jest": "^29.1.5",
-        "ts-node": "^10.9.2",
-        "typescript": "^5.5.2"
-      }
-    },
     "node_modules/@cspotcode/source-map-support": {
       "version": "0.8.1",
       "resolved": "https://registry.npmjs.org/@cspotcode/source-map-support/-/source-map-support-0.8.1.tgz",
@@ -1249,8 +1234,10 @@
       }
     },
     "node_modules/humanlayer": {
-      "resolved": "../../humanlayer-ts",
-      "link": true
+      "version": "0.7.1",
+      "resolved": "https://registry.npmjs.org/humanlayer/-/humanlayer-0.7.1.tgz",
+      "integrity": "sha512-/TffEouVEGxXBNcxC3KNMsoEsIBgUnVQzInhKRKvKLbVtbF3dJjJOVF4tojlLsUOu9vy/RDs7U3x8dtE/t4teQ==",
+      "license": "Apache-2.0"
     },
     "node_modules/iconv-lite": {
       "version": "0.4.24",