hyperbrowser 0.16.0__tar.gz → 0.17.0__tar.gz

{hyperbrowser-0.16.0 → hyperbrowser-0.17.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: hyperbrowser
-Version: 0.16.0
+Version: 0.17.0
 Summary: Python SDK for hyperbrowser
 Home-page: https://github.com/hyperbrowserai/python-sdk
 License: MIT

hyperbrowser-0.17.0/hyperbrowser/tools/__init__.py

@@ -0,0 +1,62 @@
+from hyperbrowser.models.crawl import StartCrawlJobParams
+from hyperbrowser.models.scrape import StartScrapeJobParams
+from hyperbrowser import Hyperbrowser, AsyncHyperbrowser
+
+from .openai import (
+    SCRAPE_TOOL_OPENAI,
+    CRAWL_TOOL_OPENAI,
+)
+from .anthropic import (
+    SCRAPE_TOOL_ANTHROPIC,
+    CRAWL_TOOL_ANTHROPIC,
+)
+
+
+class WebsiteScrapeTool:
+    openai_tool_definition = SCRAPE_TOOL_OPENAI
+    anthropic_tool_definition = SCRAPE_TOOL_ANTHROPIC
+
+    @staticmethod
+    def runnable(hb: Hyperbrowser, params: dict) -> str:
+        resp = hb.scrape.start_and_wait(params=StartScrapeJobParams(**params))
+        return resp.data.markdown if resp.data and resp.data.markdown else ""
+
+    @staticmethod
+    async def async_runnable(hb: AsyncHyperbrowser, params: dict) -> str:
+        resp = await hb.scrape.start_and_wait(params=StartScrapeJobParams(**params))
+        return resp.data.markdown if resp.data and resp.data.markdown else ""
+
+
+class WebsiteCrawlTool:
+    openai_tool_definition = CRAWL_TOOL_OPENAI
+    anthropic_tool_definition = CRAWL_TOOL_ANTHROPIC
+
+    @staticmethod
+    def runnable(hb: Hyperbrowser, params: dict) -> str:
+        resp = hb.crawl.start_and_wait(params=StartCrawlJobParams(**params))
+        markdown = ""
+        if resp.data:
+            for page in resp.data:
+                if page.markdown:
+                    markdown += (
+                        f"\n{'-'*50}\nUrl: {page.url}\nMarkdown:\n{page.markdown}\n"
+                    )
+        return markdown
+
+    @staticmethod
+    async def async_runnable(hb: AsyncHyperbrowser, params: dict) -> str:
+        resp = await hb.crawl.start_and_wait(params=StartCrawlJobParams(**params))
+        markdown = ""
+        if resp.data:
+            for page in resp.data:
+                if page.markdown:
+                    markdown += (
+                        f"\n{'-'*50}\nUrl: {page.url}\nMarkdown:\n{page.markdown}\n"
+                    )
+        return markdown
+
+
+__all__ = [
+    "WebsiteScrapeTool",
+    "WebsiteCrawlTool",
+]

{hyperbrowser-0.16.0 → hyperbrowser-0.17.0}/hyperbrowser/tools/openai.py

@@ -54,6 +54,7 @@ SCRAPE_TOOL_OPENAI: ChatCompletionToolParam = {
         "name": "scrape_webpage",
         "description": "Scrape content from a webpage and return the content in markdown format",
         "parameters": SCRAPE_SCHEMA,
+        "strict": True,
     },
 }
 
@@ -63,5 +64,6 @@ CRAWL_TOOL_OPENAI: ChatCompletionToolParam = {
         "name": "crawl_website",
         "description": "Crawl a website and return the content in markdown format",
         "parameters": CRAWL_SCHEMA,
+        "strict": True,
     },
 }

{hyperbrowser-0.16.0 → hyperbrowser-0.17.0}/hyperbrowser/tools/schema.py

@@ -21,6 +21,8 @@ SCRAPE_OPTIONS = {
             "description": "Whether to only return the main content of the page. If true, only the main content of the page will be returned, excluding any headers, navigation menus,footers, or other non-main content.",
         },
     },
+    "required": ["include_tags", "exclude_tags", "only_main_content"],
+    "additionalProperties": False,
 }
 
 SCRAPE_SCHEMA = {
@@ -32,7 +34,8 @@ SCRAPE_SCHEMA = {
         },
         "scrape_options": SCRAPE_OPTIONS,
     },
-    "required": ["url"],
+    "required": ["url", "scrape_options"],
+    "additionalProperties": False,
 }
 
 CRAWL_SCHEMA = {
@@ -44,7 +47,6 @@ CRAWL_SCHEMA = {
         },
         "max_pages": {
             "type": "number",
-            "default": 10,
             "description": "The maximum number of pages to crawl",
         },
         "follow_links": {
@@ -71,5 +73,14 @@ CRAWL_SCHEMA = {
         },
         "scrape_options": SCRAPE_OPTIONS,
     },
-    "required": ["url"],
+    "required": [
+        "url",
+        "max_pages",
+        "follow_links",
+        "ignore_sitemap",
+        "exclude_patterns",
+        "include_patterns",
+        "scrape_options",
+    ],
+    "additionalProperties": False,
 }

{hyperbrowser-0.16.0 → hyperbrowser-0.17.0}/pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "hyperbrowser"
-version = "0.16.0"
+version = "0.17.0"
 description = "Python SDK for hyperbrowser"
 authors = ["Nikhil Shahi <nshahi1998@gmail.com>"]
 license = "MIT"