feat(bash-py): add OpenAI and LangChain examples

bash-py/examples/langchain_agent.py

@@ -0,0 +1,72 @@
+"""
+Example: using supermemory-bash as a tool in a LangChain agent.
+
+Usage:
+    export SUPERMEMORY_API_KEY=sm-...
+    export OPENAI_API_KEY=sk-...
+    uv run --with langchain --with langchain-openai --with supermemory-bash \
+        python examples/langchain_agent.py
+"""
+
+from __future__ import annotations
+
+import asyncio
+import os
+
+from langchain.agents import create_agent
+from langchain_core.tools import tool
+from langchain_openai import ChatOpenAI
+
+from supermemory_bash import create_bash
+
+CONTAINER_TAG = os.environ.get("CONTAINER_TAG", "user_42")
+MODEL = os.environ.get("OPENAI_MODEL", "gpt-4o")
+PROMPT = os.environ.get(
+    "PROMPT",
+    "Create /notes/langchain.txt containing 'Hello from LangChain!', then cat it back to confirm.",
+)
+
+
+async def run_agent(user_message: str) -> str:
+    result = await create_bash(
+        api_key=os.environ["SUPERMEMORY_API_KEY"],
+        container_tag=CONTAINER_TAG,
+    )
+    bash = result.bash
+
+    @tool("bash", description=result.tool_description)
+    async def bash_tool(cmd: str) -> str:
+        """Execute a bash command in the Supermemory-backed shell."""
+        print(f"[bash] $ {cmd}")
+        exec_result = await bash.exec(cmd)
+        return "\n".join(
+            part
+            for part in [
+                exec_result.stdout or None,
+                f"[stderr]\n{exec_result.stderr}" if exec_result.stderr else None,
+                f"[exit {exec_result.exit_code}]",
+            ]
+            if part is not None
+        )
+
+    llm = ChatOpenAI(api_key=os.environ["OPENAI_API_KEY"], model=MODEL)
+    agent = create_agent(model=llm, tools=[bash_tool])
+
+    response = await agent.ainvoke(
+        {"messages": [{"role": "user", "content": user_message}]},
+        {"recursion_limit": 16},
+    )
+    final_message = response["messages"][-1]
+    content = getattr(final_message, "content", "")
+    if isinstance(content, str):
+        return content
+    return str(content)
+
+
+async def main() -> None:
+    answer = await run_agent(PROMPT)
+    print(f"\n[assistant] {answer}\n")
+
+
+if __name__ == "__main__":
+    asyncio.run(main())

bash-py/examples/openai_agent.py

@@ -0,0 +1,105 @@
+"""
+Example: using supermemory-bash as a tool with the OpenAI SDK.
+
+Usage:
+    export SUPERMEMORY_API_KEY=sm-...
+    export OPENAI_API_KEY=sk-...
+    uv run --with openai --with supermemory-bash python examples/openai_agent.py
+"""
+
+from __future__ import annotations
+
+import asyncio
+import json
+import os
+from typing import Any
+
+from openai import AsyncOpenAI
+
+from supermemory_bash import create_bash
+
+CONTAINER_TAG = os.environ.get("CONTAINER_TAG", "user_42")
+MODEL = os.environ.get("OPENAI_MODEL", "gpt-4o")
+PROMPT = os.environ.get(
+    "PROMPT",
+    "Create /notes/openai.txt containing 'Hello from OpenAI SDK!', then cat it back to confirm.",
+)
+
+
+async def run_agent(user_message: str) -> str:
+    result = await create_bash(
+        api_key=os.environ["SUPERMEMORY_API_KEY"],
+        container_tag=CONTAINER_TAG,
+    )
+    bash = result.bash
+
+    client = AsyncOpenAI(api_key=os.environ["OPENAI_API_KEY"])
+    tools: list[dict[str, Any]] = [
+        {
+            "type": "function",
+            "function": {
+                "name": "bash",
+                "description": result.tool_description,
+                "parameters": {
+                    "type": "object",
+                    "properties": {
+                        "cmd": {"type": "string", "description": "The bash command to execute."}
+                    },
+                    "required": ["cmd"],
+                    "additionalProperties": False,
+                },
+            },
+        }
+    ]
+
+    messages: list[dict[str, Any]] = [{"role": "user", "content": user_message}]
+
+    for _ in range(8):
+        response = await client.chat.completions.create(
+            model=MODEL,
+            messages=messages,
+            tools=tools,
+        )
+
+        message = response.choices[0].message
+        messages.append(message.model_dump(exclude_none=True))
+
+        if not message.tool_calls:
+            return message.content or ""
+
+        for tool_call in message.tool_calls:
+            if tool_call.type != "function" or tool_call.function.name != "bash":
+                continue
+
+            args = json.loads(tool_call.function.arguments or "{}")
+            cmd = args.get("cmd", "")
+            print(f"[bash] $ {cmd}")
+
+            exec_result = await bash.exec(cmd)
+            output = "\n".join(
+                part
+                for part in [
+                    exec_result.stdout or None,
+                    f"[stderr]\n{exec_result.stderr}" if exec_result.stderr else None,
+                    f"[exit {exec_result.exit_code}]",
+                ]
+                if part is not None
+            )
+            messages.append(
+                {
+                    "role": "tool",
+                    "tool_call_id": tool_call.id,
+                    "content": output,
+                }
+            )
+
+    return "(max steps reached)"
+
+
+async def main() -> None:
+    answer = await run_agent(PROMPT)
+    print(f"\n[assistant] {answer}\n")
+
+
+if __name__ == "__main__":
+    asyncio.run(main())