Add web scrapping tool

README.md

@@ -12,6 +12,7 @@ This integration provides:
 4. Selectively Enable/Disable any tool
 5. Extra LLM tools:
    * Web, maps, and news search with Duck Duck Go
+   * Web scrapping to access the Internet
    * Permanent memory tool
    * Python code execution
 
@@ -69,11 +70,12 @@ There are two options:
 
 * Extend the `custom_components.powerllm.PowerLLMTool` class to implement the functionality, then call `custom_components.powerllm.async_register_tool` to register the object of the class. See the [memory tool](https://github.com/Shulyaka/powerllm/blob/master/custom_components/powerllm/tools/memory.py) for an example
 
-* Use the `custom_components.powerllm.llm_tool` decorator for any python function. The function is recommended to have annotations for all parameters. If a parameter name is "hass", "llm_context", or any of the `homeassistant.helpers.llm.LLMContext` attributes, then the value for that parameter will be provided by the conversation agent 'pytest-style'. All other arguments will be provided by the LLM. Refer to the [python code tool](https://github.com/Shulyaka/powerllm/blob/master/custom_components/powerllm/tools/python_code.py) as an example.
+* Use the `custom_components.powerllm.llm_tool` decorator for any python function. The function is recommended to have type annotations for all parameters. If a parameter name is "hass", "llm_context", or any of the `homeassistant.helpers.llm.LLMContext` attributes, then the value for that parameter will be provided by the conversation agent ("pytest-style"). All other arguments will be provided by the LLM. Refer to the [python code tool](https://github.com/Shulyaka/powerllm/blob/master/custom_components/powerllm/tools/python_code.py) as an example.
+
+The tools in this repository use various techniques for demonstration.
 
 ## TODO
 
 * Weather forecast intent
-* Web scrapping using trafilatura
 * Ability to talk to other conversation agents (i.e. "Ask expert" for a reasoning model, or NLP conversation (Assist) for device control fallback)
 * Your suggestions!

custom_components/powerllm/__init__.py

@@ -16,6 +16,7 @@
     deferred_register_tools,
     llm_tool as llm_tool,
 )
+from .tools.web_scrape import setup as setup_web_scrape_tool
 
 _LOGGER = logging.getLogger(__name__)
 
@@ -35,5 +36,6 @@ async def async_setup(hass: HomeAssistant, config: ConfigType) -> bool:
     hass.http.register_view(LLMToolsListView)
     hass.http.register_view(LLMToolView)
     deferred_register_tools(hass)
+    setup_web_scrape_tool(hass)
 
     return True

custom_components/powerllm/manifest.json

@@ -8,6 +8,6 @@
   "integration_type": "service",
   "iot_class": "local_push",
   "issue_tracker": "https://github.com/Shulyaka/powerllm/issues",
-  "requirements": ["duckduckgo-search==6.2.9", "RestrictedPython>=7.4"],
+  "requirements": ["duckduckgo-search==6.2.9", "RestrictedPython>=7.4", "trafilatura==2.0.0"],
   "version": "0.0.1"
 }

custom_components/powerllm/tools/web_scrape.py

@@ -0,0 +1,35 @@
+"""Web scraper tool."""
+
+import json
+import logging
+
+import trafilatura
+from homeassistant.core import HomeAssistant
+
+from ..llm_tools import llm_tool
+
+_LOGGER = logging.getLogger(__name__)
+
+
+def setup(hass: HomeAssistant):
+    """Register the tool on integration startup."""
+
+    @llm_tool(hass)
+    def web_scrape(url: str):
+        """Get latest content of a web page."""
+        downloaded = trafilatura.fetch_url("linux.org.ru")
+
+        parsed = trafilatura.extract(
+            downloaded,
+            output_format="json",
+            include_links=True,
+            deduplicate=True,
+            favor_precision=True,
+        )
+
+        result = json.loads(parsed)
+
+        if "comments" in result and not result["comments"]:
+            del result["comments"]
+
+        return result

requirements_test_all.txt

@@ -21,3 +21,4 @@ ha-ffmpeg
 pymicro-vad
 duckduckgo-search
 RestrictedPython
+trafilatura

tests/conftest.py

@@ -3,7 +3,8 @@
 from unittest.mock import patch
 
 import pytest
-from homeassistant.core import HomeAssistant
+from homeassistant.core import Context, HomeAssistant
+from homeassistant.helpers import llm
 from homeassistant.setup import async_setup_component
 from pytest_homeassistant_custom_component.common import MockConfigEntry
 
@@ -60,3 +61,35 @@ async def setup_ha(hass: HomeAssistant) -> None:
     assert await async_setup_component(hass, "assist_pipeline", {})
     assert await async_setup_component(hass, "intent", {})
     assert await async_setup_component(hass, "script", {})
+
+
+@pytest.fixture
+def llm_context() -> llm.LLMContext:
+    """Return tool input context."""
+    return llm.LLMContext(
+        platform="test_platform",
+        context=Context(user_id="12345"),
+        user_prompt=None,
+        language=None,
+        assistant=None,
+        device_id=None,
+    )
+
+
+@pytest.fixture
+async def async_call_tool(
+    hass: HomeAssistant, llm_context: llm.LLMContext, mock_init_component
+):
+    """Get the tool call function."""
+
+    api = await llm.async_get_api(hass, "powerllm", llm_context)
+
+    async def _call_tool(name: str, **kwargs):
+        tool_input = llm.ToolInput(
+            tool_name=name,
+            tool_args=kwargs,
+        )
+
+        return await api.async_call_tool(tool_input)
+
+    return _call_tool

tests/test_api.py

@@ -3,7 +3,6 @@
 from decimal import Decimal
 from unittest.mock import patch
 
-import pytest
 import voluptuous as vol
 from homeassistant.components.homeassistant.exposed_entities import async_expose_entity
 from homeassistant.components.intent import async_register_timer_handler
@@ -23,19 +22,6 @@
 from custom_components.powerllm.const import CONF_PROMPT_ENTITIES
 
 
-@pytest.fixture
-def llm_context() -> llm.LLMContext:
-    """Return tool input context."""
-    return llm.LLMContext(
-        platform="test_platform",
-        context=Context(user_id="12345"),
-        user_prompt=None,
-        language=None,
-        assistant=None,
-        device_id=None,
-    )
-
-
 def test_test(hass):
     """Workaround for https://github.com/MatthewFlamm/pytest-homeassistant-custom-component/discussions/160."""
 
@@ -84,19 +70,19 @@ class MyIntentHandler(intent.IntentHandler):
 
     assert len(llm.async_get_apis(hass)) == 2
     api = await llm.async_get_api(hass, "powerllm", llm_context)
-    assert len(api.tools) == 10
+    assert len(api.tools) == 11
 
     # Match all
     intent_handler.platforms = None
 
     api = await llm.async_get_api(hass, "powerllm", llm_context)
-    assert len(api.tools) == 11
+    assert len(api.tools) == 12
 
     # Match specific domain
     intent_handler.platforms = {"light"}
 
     api = await llm.async_get_api(hass, "powerllm", llm_context)
-    assert len(api.tools) == 11
+    assert len(api.tools) == 12
     tool = api.tools[4]
     assert tool.name == "test_intent"
     assert tool.description == "Execute Home Assistant test_intent intent"
@@ -311,6 +297,7 @@ class MyIntentHandler(intent.IntentHandler):
         "maps_search",
         "memory",
         "python_code_execute",
+        "web_scrape",
     ]
 
 
@@ -327,7 +314,7 @@ class MyIntentHandler(intent.IntentHandler):
 
     assert len(llm.async_get_apis(hass)) == 2
     api = await llm.async_get_api(hass, "powerllm", llm_context)
-    assert len(api.tools) == 12
+    assert len(api.tools) == 13
     tool = api.tools[5]
     assert tool.name == "test_intent"
     assert tool.description == "my intent handler"

tests/test_llm_tools.py

@@ -1,6 +1,5 @@
 """Test powerllm config flow."""
 
-import pytest
 import voluptuous as vol
 from homeassistant.const import ATTR_FRIENDLY_NAME
 from homeassistant.core import Context, HomeAssistant, State
@@ -14,19 +13,6 @@
 from custom_components.powerllm import llm_tools
 
 
-@pytest.fixture
-def llm_context() -> llm.LLMContext:
-    """Return tool input context."""
-    return llm.LLMContext(
-        platform="test_platform",
-        context=Context(),
-        user_prompt=None,
-        language=None,
-        assistant=None,
-        device_id=None,
-    )
-
-
 def test_test(hass):
     """Workaround for https://github.com/MatthewFlamm/pytest-homeassistant-custom-component/discussions/160."""
 

tests/tools/test_python_code.py

@@ -0,0 +1,47 @@
+"""Test python script tool."""
+
+
+def test_test(hass):
+    """Workaround for https://github.com/MatthewFlamm/pytest-homeassistant-custom-component/discussions/160."""
+
+
+async def test_python_script_tool(async_call_tool) -> None:
+    """Test python script tool."""
+
+    source = """
+output["test"] = "passed"
+output["test2"] = "passed2"
+    """
+
+    response = await async_call_tool("python_code_execute", source=source)
+
+    assert response == {"output": {"test": "passed", "test2": "passed2"}}
+
+
+async def test_python_script_tool_import(async_call_tool) -> None:
+    """Test python script tool with import."""
+
+    source = """
+import math
+output["test"] = math.cos(0)
+    """
+
+    response = await async_call_tool("python_code_execute", source=source)
+
+    assert response == {"output": {"test": 1.0}}
+
+
+async def test_python_script_tool_print(async_call_tool) -> None:
+    """Test print in python script tool."""
+
+    source = """
+print("test1")
+def test2():
+    print("test2")
+
+test2()
+    """
+
+    response = await async_call_tool("python_code_execute", source=source)
+
+    assert response == {"printed": "test1\ntest2\n"}

tests/tools/test_web_scrape.py

@@ -0,0 +1,30 @@
+"""Test web scrape tool."""
+
+from unittest.mock import patch
+
+
+def test_test(hass):
+    """Workaround for https://github.com/MatthewFlamm/pytest-homeassistant-custom-component/discussions/160."""
+
+
+async def test_web_scrape_tool(async_call_tool) -> None:
+    """Test web scrape tool."""
+
+    helloworld = """<!DOCTYPE html>
+<html lang="en">
+<head>
+    <meta charset="UTF-8">
+    <meta name="viewport" content="width=device-width, initial-scale=1.0">
+    <title>Hello World</title>
+</head>
+<body>
+    <h1>Hello, World!</h1>
+    <p>This is a simple HTML page with a greeting message.</p>
+</body>
+</html>
+"""
+
+    with patch("trafilatura.fetch_url", return_value=helloworld):
+        response = await async_call_tool("web_scrape", url="example.com")
+
+    assert response == {"text": "This is a simple HTML page with a greeting message."}