hyperbrowser 0.33.0__tar.gz → 0.34.0__tar.gz

{hyperbrowser-0.33.0 → hyperbrowser-0.34.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: hyperbrowser
-Version: 0.33.0
+Version: 0.34.0
 Summary: Python SDK for hyperbrowser
 License: MIT
 Author: Nikhil Shahi
@@ -15,6 +15,7 @@ Classifier: Programming Language :: Python :: 3.11
 Classifier: Programming Language :: Python :: 3.12
 Classifier: Programming Language :: Python :: 3.13
 Requires-Dist: httpx (>=0.23.0,<1)
+Requires-Dist: jsonref (>=1.1.0)
 Requires-Dist: pydantic (>=2.0,<3)
 Project-URL: Homepage, https://github.com/hyperbrowserai/python-sdk
 Project-URL: Repository, https://github.com/hyperbrowserai/python-sdk

{hyperbrowser-0.33.0 → hyperbrowser-0.34.0}/hyperbrowser/client/async_client.py

@@ -1,15 +1,15 @@
 from typing import Optional
 
+from ..config import ClientConfig
+from ..transport.async_transport import AsyncTransport
+from .base import HyperbrowserBase
+from .managers.async_manager.agents import Agents
+from .managers.async_manager.crawl import CrawlManager
+from .managers.async_manager.extension import ExtensionManager
 from .managers.async_manager.extract import ExtractManager
 from .managers.async_manager.profile import ProfileManager
-from .managers.async_manager.session import SessionManager
 from .managers.async_manager.scrape import ScrapeManager
-from .managers.async_manager.crawl import CrawlManager
-from .managers.async_manager.extension import ExtensionManager
-from .managers.async_manager.beta import Beta
-from .base import HyperbrowserBase
-from ..transport.async_transport import AsyncTransport
-from ..config import ClientConfig
+from .managers.async_manager.session import SessionManager
 
 
 class AsyncHyperbrowser(HyperbrowserBase):
@@ -30,7 +30,7 @@ class AsyncHyperbrowser(HyperbrowserBase):
         self.extract = ExtractManager(self)
         self.profiles = ProfileManager(self)
         self.extensions = ExtensionManager(self)
-        self.beta = Beta(self)
+        self.agents = Agents(self)
 
     async def close(self) -> None:
         await self.transport.close()

{hyperbrowser-0.33.0/hyperbrowser/client/managers/async_manager/beta → hyperbrowser-0.34.0/hyperbrowser/client/managers/async_manager}/agents/__init__.py

@@ -1,5 +1,6 @@
+from .browser_use import BrowserUseManager
+
+
 class Agents:
     def __init__(self, client):
-        from .browser_use import BrowserUseManager
-
         self.browser_use = BrowserUseManager(client)

{hyperbrowser-0.33.0/hyperbrowser/client/managers/async_manager/beta → hyperbrowser-0.34.0/hyperbrowser/client/managers/async_manager}/agents/browser_use.py

@@ -1,12 +1,14 @@
 import asyncio
+
 from hyperbrowser.exceptions import HyperbrowserError
-from ......models import (
+
+from .....models import (
     POLLING_ATTEMPTS,
     BasicResponse,
+    BrowserUseTaskResponse,
+    BrowserUseTaskStatusResponse,
     StartBrowserUseTaskParams,
     StartBrowserUseTaskResponse,
-    BrowserUseTaskStatusResponse,
-    BrowserUseTaskResponse,
 )
 
 

{hyperbrowser-0.33.0 → hyperbrowser-0.34.0}/hyperbrowser/client/managers/async_manager/crawl.py

@@ -3,6 +3,8 @@ import asyncio
 from hyperbrowser.models.consts import POLLING_ATTEMPTS
 from ....models.crawl import (
     CrawlJobResponse,
+    CrawlJobStatus,
+    CrawlJobStatusResponse,
     GetCrawlJobParams,
     StartCrawlJobParams,
     StartCrawlJobResponse,
@@ -21,11 +23,18 @@ class CrawlManager:
         )
         return StartCrawlJobResponse(**response.data)
 
+    async def get_status(self, job_id: str) -> CrawlJobStatusResponse:
+        response = await self._client.transport.get(
+            self._client._build_url(f"/crawl/{job_id}/status")
+        )
+        return CrawlJobStatusResponse(**response.data)
+
     async def get(
         self, job_id: str, params: GetCrawlJobParams = GetCrawlJobParams()
     ) -> CrawlJobResponse:
         response = await self._client.transport.get(
-            self._client._build_url(f"/crawl/{job_id}"), params=params.__dict__
+            self._client._build_url(f"/crawl/{job_id}"),
+            params=params.model_dump(exclude_none=True, by_alias=True),
         )
         return CrawlJobResponse(**response.data)
 
@@ -37,18 +46,13 @@ class CrawlManager:
         if not job_id:
             raise HyperbrowserError("Failed to start crawl job")
 
-        job_response: CrawlJobResponse
+        job_status: CrawlJobStatus = "pending"
         failures = 0
         while True:
             try:
-                job_response = await self.get(
-                    job_id,
-                    params=GetCrawlJobParams(batch_size=1),
-                )
-                if (
-                    job_response.status == "completed"
-                    or job_response.status == "failed"
-                ):
+                job_status_resp = await self.get_status(job_id)
+                job_status = job_status_resp.status
+                if job_status == "completed" or job_status == "failed":
                     break
             except Exception as e:
                 failures += 1
@@ -62,8 +66,7 @@ class CrawlManager:
         if not return_all_pages:
             while True:
                 try:
-                    job_response = await self.get(job_id)
-                    return job_response
+                    return await self.get(job_id)
                 except Exception as e:
                     failures += 1
                     if failures >= POLLING_ATTEMPTS:
@@ -73,9 +76,20 @@ class CrawlManager:
                 await asyncio.sleep(0.5)
 
         failures = 0
-        job_response.current_page_batch = 0
-        job_response.data = []
-        while job_response.current_page_batch < job_response.total_page_batches:
+        job_response = CrawlJobResponse(
+            jobId=job_id,
+            status=job_status,
+            data=[],
+            currentPageBatch=0,
+            totalPageBatches=0,
+            totalCrawledPages=0,
+            batchSize=100,
+        )
+        first_check = True
+        while (
+            first_check
+            or job_response.current_page_batch < job_response.total_page_batches
+        ):
             try:
                 tmp_job_response = await self.get(
                     job_start_resp.job_id,
@@ -90,6 +104,7 @@ class CrawlManager:
                 job_response.total_page_batches = tmp_job_response.total_page_batches
                 job_response.batch_size = tmp_job_response.batch_size
                 failures = 0
+                first_check = False
             except Exception as e:
                 failures += 1
                 if failures >= POLLING_ATTEMPTS:

{hyperbrowser-0.33.0 → hyperbrowser-0.34.0}/hyperbrowser/client/managers/async_manager/extract.py

@@ -3,9 +3,11 @@ from hyperbrowser.exceptions import HyperbrowserError
 from hyperbrowser.models.consts import POLLING_ATTEMPTS
 from hyperbrowser.models.extract import (
     ExtractJobResponse,
+    ExtractJobStatusResponse,
     StartExtractJobParams,
     StartExtractJobResponse,
 )
+import jsonref
 
 
 class ExtractManager:
@@ -17,7 +19,9 @@ class ExtractManager:
             raise HyperbrowserError("Either schema or prompt must be provided")
         if params.schema_:
             if hasattr(params.schema_, "model_json_schema"):
-                params.schema_ = params.schema_.model_json_schema()
+                params.schema_ = jsonref.replace_refs(
+                    params.schema_.model_json_schema(), proxies=False, lazy_load=False
+                )
 
         response = await self._client.transport.post(
             self._client._build_url("/extract"),
@@ -25,6 +29,12 @@ class ExtractManager:
         )
         return StartExtractJobResponse(**response.data)
 
+    async def get_status(self, job_id: str) -> ExtractJobStatusResponse:
+        response = await self._client.transport.get(
+            self._client._build_url(f"/extract/{job_id}/status")
+        )
+        return ExtractJobStatusResponse(**response.data)
+
     async def get(self, job_id: str) -> ExtractJobResponse:
         response = await self._client.transport.get(
             self._client._build_url(f"/extract/{job_id}")
@@ -40,12 +50,10 @@ class ExtractManager:
         failures = 0
         while True:
             try:
-                job_response = await self.get(job_id)
-                if (
-                    job_response.status == "completed"
-                    or job_response.status == "failed"
-                ):
-                    return job_response
+                job_status_resp = await self.get_status(job_id)
+                job_status = job_status_resp.status
+                if job_status == "completed" or job_status == "failed":
+                    return await self.get(job_id)
                 failures = 0
             except Exception as e:
                 failures += 1

{hyperbrowser-0.33.0 → hyperbrowser-0.34.0}/hyperbrowser/client/managers/async_manager/profile.py

@@ -33,6 +33,7 @@ class ProfileManager:
         self, params: ProfileListParams = ProfileListParams()
     ) -> ProfileListResponse:
         response = await self._client.transport.get(
-            self._client._build_url("/profiles"), params=params.__dict__
+            self._client._build_url("/profiles"),
+            params=params.model_dump(exclude_none=True, by_alias=True),
         )
         return ProfileListResponse(**response.data)

{hyperbrowser-0.33.0 → hyperbrowser-0.34.0}/hyperbrowser/client/managers/async_manager/scrape.py

@@ -4,8 +4,11 @@ from typing import Optional
 from hyperbrowser.models.consts import POLLING_ATTEMPTS
 from ....models.scrape import (
     BatchScrapeJobResponse,
+    BatchScrapeJobStatusResponse,
     GetBatchScrapeJobParams,
     ScrapeJobResponse,
+    ScrapeJobStatus,
+    ScrapeJobStatusResponse,
     StartBatchScrapeJobParams,
     StartBatchScrapeJobResponse,
     StartScrapeJobParams,
@@ -27,11 +30,18 @@ class BatchScrapeManager:
         )
         return StartBatchScrapeJobResponse(**response.data)
 
+    async def get_status(self, job_id: str) -> BatchScrapeJobStatusResponse:
+        response = await self._client.transport.get(
+            self._client._build_url(f"/scrape/batch/{job_id}/status")
+        )
+        return BatchScrapeJobStatusResponse(**response.data)
+
     async def get(
         self, job_id: str, params: GetBatchScrapeJobParams = GetBatchScrapeJobParams()
     ) -> BatchScrapeJobResponse:
         response = await self._client.transport.get(
-            self._client._build_url(f"/scrape/batch/{job_id}"), params=params.__dict__
+            self._client._build_url(f"/scrape/batch/{job_id}"),
+            params=params.model_dump(exclude_none=True, by_alias=True),
         )
         return BatchScrapeJobResponse(**response.data)
 
@@ -43,19 +53,14 @@ class BatchScrapeManager:
         if not job_id:
             raise HyperbrowserError("Failed to start batch scrape job")
 
-        job_response: BatchScrapeJobResponse
+        job_status: ScrapeJobStatus = "pending"
         failures = 0
         while True:
             try:
-                job_response = await self.get(
-                    job_id, params=GetBatchScrapeJobParams(batch_size=1)
-                )
-                if (
-                    job_response.status == "completed"
-                    or job_response.status == "failed"
-                ):
+                job_status_resp = await self.get_status(job_id)
+                job_status = job_status_resp.status
+                if job_status == "completed" or job_status == "failed":
                     break
-                failures = 0
             except Exception as e:
                 failures += 1
                 if failures >= POLLING_ATTEMPTS:
@@ -68,8 +73,7 @@ class BatchScrapeManager:
         if not return_all_pages:
             while True:
                 try:
-                    job_response = await self.get(job_id)
-                    return job_response
+                    return await self.get(job_id)
                 except Exception as e:
                     failures += 1
                     if failures >= POLLING_ATTEMPTS:
@@ -79,9 +83,21 @@ class BatchScrapeManager:
                 await asyncio.sleep(0.5)
 
         failures = 0
-        job_response.current_page_batch = 0
-        job_response.data = []
-        while job_response.current_page_batch < job_response.total_page_batches:
+        job_response = BatchScrapeJobResponse(
+            jobId=job_id,
+            status=job_status,
+            data=[],
+            currentPageBatch=0,
+            totalPageBatches=0,
+            totalScrapedPages=0,
+            batchSize=100,
+        )
+        first_check = True
+
+        while (
+            first_check
+            or job_response.current_page_batch < job_response.total_page_batches
+        ):
             try:
                 tmp_job_response = await self.get(
                     job_id,
@@ -96,6 +112,7 @@ class BatchScrapeManager:
                 job_response.total_page_batches = tmp_job_response.total_page_batches
                 job_response.batch_size = tmp_job_response.batch_size
                 failures = 0
+                first_check = False
             except Exception as e:
                 failures += 1
                 if failures >= POLLING_ATTEMPTS:
@@ -119,6 +136,12 @@ class ScrapeManager:
         )
         return StartScrapeJobResponse(**response.data)
 
+    async def get_status(self, job_id: str) -> ScrapeJobStatusResponse:
+        response = await self._client.transport.get(
+            self._client._build_url(f"/scrape/{job_id}/status")
+        )
+        return ScrapeJobStatusResponse(**response.data)
+
     async def get(self, job_id: str) -> ScrapeJobResponse:
         response = await self._client.transport.get(
             self._client._build_url(f"/scrape/{job_id}")
@@ -134,12 +157,10 @@ class ScrapeManager:
         failures = 0
         while True:
             try:
-                job_response = await self.get(job_id)
-                if (
-                    job_response.status == "completed"
-                    or job_response.status == "failed"
-                ):
-                    return job_response
+                job_status_resp = await self.get_status(job_id)
+                job_status = job_status_resp.status
+                if job_status == "completed" or job_status == "failed":
+                    return await self.get(job_id)
                 failures = 0
             except Exception as e:
                 failures += 1

{hyperbrowser-0.33.0 → hyperbrowser-0.34.0}/hyperbrowser/client/managers/async_manager/session.py

@@ -42,7 +42,8 @@ class SessionManager:
         self, params: SessionListParams = SessionListParams()
     ) -> SessionListResponse:
         response = await self._client.transport.get(
-            self._client._build_url("/sessions"), params=params.__dict__
+            self._client._build_url("/sessions"),
+            params=params.model_dump(exclude_none=True, by_alias=True),
         )
         return SessionListResponse(**response.data)
 

{hyperbrowser-0.33.0/hyperbrowser/client/managers/sync_manager/beta → hyperbrowser-0.34.0/hyperbrowser/client/managers/sync_manager}/agents/__init__.py

@@ -1,5 +1,6 @@
+from .browser_use import BrowserUseManager
+
+
 class Agents:
     def __init__(self, client):
-        from .browser_use import BrowserUseManager
-
         self.browser_use = BrowserUseManager(client)

{hyperbrowser-0.33.0/hyperbrowser/client/managers/sync_manager/beta → hyperbrowser-0.34.0/hyperbrowser/client/managers/sync_manager}/agents/browser_use.py

@@ -1,12 +1,14 @@
 import time
+
 from hyperbrowser.exceptions import HyperbrowserError
-from ......models import (
+
+from .....models import (
     POLLING_ATTEMPTS,
     BasicResponse,
+    BrowserUseTaskResponse,
+    BrowserUseTaskStatusResponse,
     StartBrowserUseTaskParams,
     StartBrowserUseTaskResponse,
-    BrowserUseTaskStatusResponse,
-    BrowserUseTaskResponse,
 )
 
 

{hyperbrowser-0.33.0 → hyperbrowser-0.34.0}/hyperbrowser/client/managers/sync_manager/crawl.py

@@ -4,6 +4,8 @@ from typing import Optional
 from hyperbrowser.models.consts import POLLING_ATTEMPTS
 from ....models.crawl import (
     CrawlJobResponse,
+    CrawlJobStatus,
+    CrawlJobStatusResponse,
     GetCrawlJobParams,
     StartCrawlJobParams,
     StartCrawlJobResponse,
@@ -22,11 +24,18 @@ class CrawlManager:
         )
         return StartCrawlJobResponse(**response.data)
 
+    def get_status(self, job_id: str) -> CrawlJobStatusResponse:
+        response = self._client.transport.get(
+            self._client._build_url(f"/crawl/{job_id}/status")
+        )
+        return CrawlJobStatusResponse(**response.data)
+
     def get(
         self, job_id: str, params: GetCrawlJobParams = GetCrawlJobParams()
     ) -> CrawlJobResponse:
         response = self._client.transport.get(
-            self._client._build_url(f"/crawl/{job_id}"), params=params.__dict__
+            self._client._build_url(f"/crawl/{job_id}"),
+            params=params.model_dump(exclude_none=True, by_alias=True),
         )
         return CrawlJobResponse(**response.data)
 
@@ -38,18 +47,13 @@ class CrawlManager:
         if not job_id:
             raise HyperbrowserError("Failed to start crawl job")
 
-        job_response: CrawlJobResponse
+        job_status: CrawlJobStatus = "pending"
         failures = 0
         while True:
             try:
-                job_response = self.get(
-                    job_id,
-                    params=GetCrawlJobParams(batch_size=1),
-                )
-                if (
-                    job_response.status == "completed"
-                    or job_response.status == "failed"
-                ):
+                job_status_resp = self.get_status(job_id)
+                job_status = job_status_resp.status
+                if job_status == "completed" or job_status == "failed":
                     break
             except Exception as e:
                 failures += 1
@@ -63,8 +67,7 @@ class CrawlManager:
         if not return_all_pages:
             while True:
                 try:
-                    job_response = self.get(job_id)
-                    return job_response
+                    return self.get(job_id)
                 except Exception as e:
                     failures += 1
                     if failures >= POLLING_ATTEMPTS:
@@ -74,12 +77,23 @@ class CrawlManager:
                 time.sleep(0.5)
 
         failures = 0
-        job_response.current_page_batch = 0
-        job_response.data = []
-        while job_response.current_page_batch < job_response.total_page_batches:
+        job_response = CrawlJobResponse(
+            jobId=job_id,
+            status=job_status,
+            data=[],
+            currentPageBatch=0,
+            totalPageBatches=0,
+            totalCrawledPages=0,
+            batchSize=100,
+        )
+        first_check = True
+        while (
+            first_check
+            or job_response.current_page_batch < job_response.total_page_batches
+        ):
             try:
                 tmp_job_response = self.get(
-                    job_id,
+                    job_start_resp.job_id,
                     GetCrawlJobParams(
                         page=job_response.current_page_batch + 1, batch_size=100
                     ),
@@ -91,6 +105,7 @@ class CrawlManager:
                 job_response.total_page_batches = tmp_job_response.total_page_batches
                 job_response.batch_size = tmp_job_response.batch_size
                 failures = 0
+                first_check = False
             except Exception as e:
                 failures += 1
                 if failures >= POLLING_ATTEMPTS:

{hyperbrowser-0.33.0 → hyperbrowser-0.34.0}/hyperbrowser/client/managers/sync_manager/extract.py

@@ -3,9 +3,11 @@ from hyperbrowser.exceptions import HyperbrowserError
 from hyperbrowser.models.consts import POLLING_ATTEMPTS
 from hyperbrowser.models.extract import (
     ExtractJobResponse,
+    ExtractJobStatusResponse,
     StartExtractJobParams,
     StartExtractJobResponse,
 )
+import jsonref
 
 
 class ExtractManager:
@@ -17,7 +19,9 @@ class ExtractManager:
             raise HyperbrowserError("Either schema or prompt must be provided")
         if params.schema_:
             if hasattr(params.schema_, "model_json_schema"):
-                params.schema_ = params.schema_.model_json_schema()
+                params.schema_ = jsonref.replace_refs(
+                    params.schema_.model_json_schema(), proxies=False, lazy_load=False
+                )
 
         response = self._client.transport.post(
             self._client._build_url("/extract"),
@@ -25,6 +29,12 @@ class ExtractManager:
         )
         return StartExtractJobResponse(**response.data)
 
+    def get_status(self, job_id: str) -> ExtractJobStatusResponse:
+        response = self._client.transport.get(
+            self._client._build_url(f"/extract/{job_id}/status")
+        )
+        return ExtractJobStatusResponse(**response.data)
+
     def get(self, job_id: str) -> ExtractJobResponse:
         response = self._client.transport.get(
             self._client._build_url(f"/extract/{job_id}")
@@ -40,12 +50,10 @@ class ExtractManager:
         failures = 0
         while True:
             try:
-                job_response = self.get(job_start_resp.job_id)
-                if (
-                    job_response.status == "completed"
-                    or job_response.status == "failed"
-                ):
-                    return job_response
+                job_status_resp = self.get_status(job_id)
+                job_status = job_status_resp.status
+                if job_status == "completed" or job_status == "failed":
+                    return self.get(job_id)
             except Exception as e:
                 failures += 1
                 if failures >= POLLING_ATTEMPTS:

{hyperbrowser-0.33.0 → hyperbrowser-0.34.0}/hyperbrowser/client/managers/sync_manager/profile.py

@@ -33,6 +33,7 @@ class ProfileManager:
         self, params: ProfileListParams = ProfileListParams()
     ) -> ProfileListResponse:
         response = self._client.transport.get(
-            self._client._build_url("/profiles"), params=params.__dict__
+            self._client._build_url("/profiles"),
+            params=params.model_dump(exclude_none=True, by_alias=True),
         )
         return ProfileListResponse(**response.data)

{hyperbrowser-0.33.0 → hyperbrowser-0.34.0}/hyperbrowser/client/managers/sync_manager/scrape.py

@@ -4,8 +4,11 @@ from typing import Optional
 from hyperbrowser.models.consts import POLLING_ATTEMPTS
 from ....models.scrape import (
     BatchScrapeJobResponse,
+    BatchScrapeJobStatusResponse,
     GetBatchScrapeJobParams,
     ScrapeJobResponse,
+    ScrapeJobStatus,
+    ScrapeJobStatusResponse,
     StartBatchScrapeJobParams,
     StartBatchScrapeJobResponse,
     StartScrapeJobParams,
@@ -25,11 +28,18 @@ class BatchScrapeManager:
         )
         return StartBatchScrapeJobResponse(**response.data)
 
+    def get_status(self, job_id: str) -> BatchScrapeJobStatusResponse:
+        response = self._client.transport.get(
+            self._client._build_url(f"/scrape/batch/{job_id}/status")
+        )
+        return BatchScrapeJobStatusResponse(**response.data)
+
     def get(
         self, job_id: str, params: GetBatchScrapeJobParams = GetBatchScrapeJobParams()
     ) -> BatchScrapeJobResponse:
         response = self._client.transport.get(
-            self._client._build_url(f"/scrape/batch/{job_id}"), params=params.__dict__
+            self._client._build_url(f"/scrape/batch/{job_id}"),
+            params=params.model_dump(exclude_none=True, by_alias=True),
         )
         return BatchScrapeJobResponse(**response.data)
 
@@ -41,19 +51,14 @@ class BatchScrapeManager:
         if not job_id:
             raise HyperbrowserError("Failed to start batch scrape job")
 
-        job_response: BatchScrapeJobResponse
+        job_status: ScrapeJobStatus = "pending"
         failures = 0
         while True:
             try:
-                job_response = self.get(
-                    job_id, params=GetBatchScrapeJobParams(batch_size=1)
-                )
-                if (
-                    job_response.status == "completed"
-                    or job_response.status == "failed"
-                ):
+                job_status_resp = self.get_status(job_id)
+                job_status = job_status_resp.status
+                if job_status == "completed" or job_status == "failed":
                     break
-                failures = 0
             except Exception as e:
                 failures += 1
                 if failures >= POLLING_ATTEMPTS:
@@ -66,8 +71,7 @@ class BatchScrapeManager:
         if not return_all_pages:
             while True:
                 try:
-                    job_response = self.get(job_id)
-                    return job_response
+                    return self.get(job_id)
                 except Exception as e:
                     failures += 1
                     if failures >= POLLING_ATTEMPTS:
@@ -77,13 +81,25 @@ class BatchScrapeManager:
                 time.sleep(0.5)
 
         failures = 0
-        job_response.current_page_batch = 0
-        job_response.data = []
-        while job_response.current_page_batch < job_response.total_page_batches:
+        job_response = BatchScrapeJobResponse(
+            jobId=job_id,
+            status=job_status,
+            data=[],
+            currentPageBatch=0,
+            totalPageBatches=0,
+            totalScrapedPages=0,
+            batchSize=100,
+        )
+        first_check = True
+
+        while (
+            first_check
+            or job_response.current_page_batch < job_response.total_page_batches
+        ):
             try:
                 tmp_job_response = self.get(
-                    job_start_resp.job_id,
-                    GetBatchScrapeJobParams(
+                    job_id,
+                    params=GetBatchScrapeJobParams(
                         page=job_response.current_page_batch + 1, batch_size=100
                     ),
                 )
@@ -94,6 +110,7 @@ class BatchScrapeManager:
                 job_response.total_page_batches = tmp_job_response.total_page_batches
                 job_response.batch_size = tmp_job_response.batch_size
                 failures = 0
+                first_check = False
             except Exception as e:
                 failures += 1
                 if failures >= POLLING_ATTEMPTS:
@@ -117,6 +134,12 @@ class ScrapeManager:
         )
         return StartScrapeJobResponse(**response.data)
 
+    def get_status(self, job_id: str) -> ScrapeJobStatusResponse:
+        response = self._client.transport.get(
+            self._client._build_url(f"/scrape/{job_id}/status")
+        )
+        return ScrapeJobStatusResponse(**response.data)
+
     def get(self, job_id: str) -> ScrapeJobResponse:
         response = self._client.transport.get(
             self._client._build_url(f"/scrape/{job_id}")
@@ -132,12 +155,10 @@ class ScrapeManager:
         failures = 0
         while True:
             try:
-                job_response = self.get(job_id)
-                if (
-                    job_response.status == "completed"
-                    or job_response.status == "failed"
-                ):
-                    return job_response
+                job_status_resp = self.get_status(job_id)
+                job_status = job_status_resp.status
+                if job_status == "completed" or job_status == "failed":
+                    return self.get(job_id)
                 failures = 0
             except Exception as e:
                 failures += 1

{hyperbrowser-0.33.0 → hyperbrowser-0.34.0}/hyperbrowser/client/managers/sync_manager/session.py

@@ -40,7 +40,8 @@ class SessionManager:
         self, params: SessionListParams = SessionListParams()
     ) -> SessionListResponse:
         response = self._client.transport.get(
-            self._client._build_url("/sessions"), params=params.__dict__
+            self._client._build_url("/sessions"),
+            params=params.model_dump(exclude_none=True, by_alias=True),
         )
         return SessionListResponse(**response.data)
 

{hyperbrowser-0.33.0 → hyperbrowser-0.34.0}/hyperbrowser/client/sync.py

@@ -1,15 +1,15 @@
 from typing import Optional
 
+from ..config import ClientConfig
+from ..transport.sync import SyncTransport
+from .base import HyperbrowserBase
+from .managers.sync_manager.agents import Agents
+from .managers.sync_manager.crawl import CrawlManager
+from .managers.sync_manager.extension import ExtensionManager
 from .managers.sync_manager.extract import ExtractManager
 from .managers.sync_manager.profile import ProfileManager
-from .managers.sync_manager.session import SessionManager
 from .managers.sync_manager.scrape import ScrapeManager
-from .managers.sync_manager.crawl import CrawlManager
-from .managers.sync_manager.beta import Beta
-from .managers.sync_manager.extension import ExtensionManager
-from .base import HyperbrowserBase
-from ..transport.sync import SyncTransport
-from ..config import ClientConfig
+from .managers.sync_manager.session import SessionManager
 
 
 class Hyperbrowser(HyperbrowserBase):
@@ -30,7 +30,7 @@ class Hyperbrowser(HyperbrowserBase):
         self.extract = ExtractManager(self)
         self.profiles = ProfileManager(self)
         self.extensions = ExtensionManager(self)
-        self.beta = Beta(self)
+        self.agents = Agents(self)
 
     def close(self) -> None:
         self.transport.close()

{hyperbrowser-0.33.0 → hyperbrowser-0.34.0}/hyperbrowser/models/__init__.py

@@ -1,142 +1,151 @@
+from .agents.browser_use import (
+    BrowserUseTaskData,
+    BrowserUseTaskResponse,
+    BrowserUseTaskStatusResponse,
+    StartBrowserUseTaskParams,
+    StartBrowserUseTaskResponse,
+)
 from .consts import (
-    ScrapeFormat,
-    ScrapeWaitUntil,
-    ScrapePageStatus,
-    ScrapeScreenshotFormat,
-    RecordingStatus,
-    DownloadsStatus,
+    ISO639_1,
     POLLING_ATTEMPTS,
+    BrowserUseLlm,
     Country,
+    DownloadsStatus,
     OperatingSystem,
     Platform,
-    ISO639_1,
+    RecordingStatus,
+    ScrapeFormat,
+    ScrapePageStatus,
+    ScrapeScreenshotFormat,
+    ScrapeWaitUntil,
     State,
-    BrowserUseLlm,
 )
 from .crawl import (
+    CrawledPage,
+    CrawlJobResponse,
     CrawlJobStatus,
+    CrawlJobStatusResponse,
     CrawlPageStatus,
+    GetCrawlJobParams,
     StartCrawlJobParams,
     StartCrawlJobResponse,
-    CrawledPage,
-    GetCrawlJobParams,
-    CrawlJobResponse,
 )
 from .extension import CreateExtensionParams, ExtensionResponse
 from .extract import (
+    ExtractJobResponse,
     ExtractJobStatus,
+    ExtractJobStatusResponse,
     StartExtractJobParams,
     StartExtractJobResponse,
-    ExtractJobResponse,
 )
 from .profile import (
     CreateProfileResponse,
-    ProfileResponse,
     ProfileListParams,
     ProfileListResponse,
+    ProfileResponse,
 )
 from .scrape import (
-    ScrapeJobStatus,
-    ScreenshotOptions,
-    ScrapeOptions,
-    StartScrapeJobParams,
-    StartScrapeJobResponse,
+    BatchScrapeJobResponse,
+    BatchScrapeJobStatusResponse,
+    GetBatchScrapeJobParams,
+    ScrapedPage,
     ScrapeJobData,
     ScrapeJobResponse,
+    ScrapeJobStatus,
+    ScrapeJobStatusResponse,
+    ScrapeOptions,
+    ScreenshotOptions,
     StartBatchScrapeJobParams,
-    ScrapedPage,
-    GetBatchScrapeJobParams,
     StartBatchScrapeJobResponse,
-    BatchScrapeJobResponse,
+    StartScrapeJobParams,
+    StartScrapeJobResponse,
 )
 from .session import (
-    SessionStatus,
     BasicResponse,
+    CreateSessionParams,
+    CreateSessionProfile,
+    GetSessionDownloadsUrlResponse,
+    GetSessionRecordingUrlResponse,
+    ScreenConfig,
     Session,
     SessionDetail,
     SessionListParams,
     SessionListResponse,
-    ScreenConfig,
-    CreateSessionProfile,
-    CreateSessionParams,
     SessionRecording,
-    GetSessionRecordingUrlResponse,
-    GetSessionDownloadsUrlResponse,
-)
-from .beta.agents.browser_use import (
-    StartBrowserUseTaskParams,
-    StartBrowserUseTaskResponse,
-    BrowserUseTaskStatusResponse,
-    BrowserUseTaskData,
-    BrowserUseTaskResponse,
+    SessionStatus,
 )
 
 __all__ = [
     # consts
-    "BrowserUseLlm",
-    "ScrapeFormat",
-    "ScrapeWaitUntil",
-    "ScrapePageStatus",
-    "ScrapeScreenshotFormat",
-    "RecordingStatus",
-    "DownloadsStatus",
+    "ISO639_1",
     "POLLING_ATTEMPTS",
+    "BrowserUseLlm",
     "Country",
+    "DownloadsStatus",
     "OperatingSystem",
     "Platform",
-    "ISO639_1",
+    "RecordingStatus",
+    "ScrapeFormat",
+    "ScrapePageStatus",
+    "ScrapeScreenshotFormat",
+    "ScrapeWaitUntil",
     "State",
+    # agents
+    "BrowserUseTaskStatus",
+    "BrowserUseTaskData",
+    "BrowserUseTaskResponse",
+    "BrowserUseTaskStatusResponse",
+    "StartBrowserUseTaskParams",
+    "StartBrowserUseTaskResponse",
     # crawl
+    "CrawledPage",
+    "CrawlJobResponse",
     "CrawlJobStatus",
+    "CrawlJobStatusResponse",
     "CrawlPageStatus",
+    "GetCrawlJobParams",
     "StartCrawlJobParams",
     "StartCrawlJobResponse",
-    "CrawledPage",
-    "GetCrawlJobParams",
-    "CrawlJobResponse",
     # extension
     "CreateExtensionParams",
     "ExtensionResponse",
     # extract
+    "ExtractJobResponse",
     "ExtractJobStatus",
+    "ExtractJobStatusResponse",
     "StartExtractJobParams",
     "StartExtractJobResponse",
-    "ExtractJobResponse",
     # profile
     "CreateProfileResponse",
-    "ProfileResponse",
     "ProfileListParams",
     "ProfileListResponse",
+    "ProfileResponse",
     # scrape
-    "ScrapeJobStatus",
-    "ScreenshotOptions",
-    "ScrapeOptions",
-    "StartScrapeJobParams",
-    "StartScrapeJobResponse",
+    "BatchScrapeJobResponse",
+    "BatchScrapeJobStatusResponse",
+    "GetBatchScrapeJobParams",
+    "ScrapedPage",
     "ScrapeJobData",
     "ScrapeJobResponse",
+    "ScrapeJobStatus",
+    "ScrapeJobStatusResponse",
+    "ScrapeOptions",
+    "ScreenshotOptions",
     "StartBatchScrapeJobParams",
-    "ScrapedPage",
-    "GetBatchScrapeJobParams",
     "StartBatchScrapeJobResponse",
-    "BatchScrapeJobResponse",
+    "StartScrapeJobParams",
+    "StartScrapeJobResponse",
     # session
-    "SessionStatus",
     "BasicResponse",
+    "CreateSessionParams",
+    "CreateSessionProfile",
+    "GetSessionDownloadsUrlResponse",
+    "GetSessionRecordingUrlResponse",
+    "ScreenConfig",
     "Session",
     "SessionDetail",
     "SessionListParams",
     "SessionListResponse",
-    "ScreenConfig",
-    "CreateSessionProfile",
-    "CreateSessionParams",
     "SessionRecording",
-    "GetSessionRecordingUrlResponse",
-    "GetSessionDownloadsUrlResponse",
-    # agents
-    "StartBrowserUseTaskParams",
-    "StartBrowserUseTaskResponse",
-    "BrowserUseTaskStatusResponse",
-    "BrowserUseTaskData",
-    "BrowserUseTaskResponse",
+    "SessionStatus",
 ]

{hyperbrowser-0.33.0/hyperbrowser/models/beta → hyperbrowser-0.34.0/hyperbrowser/models}/agents/browser_use.py

@@ -1,8 +1,9 @@
 from typing import Literal, Optional, Union
+
 from pydantic import BaseModel, ConfigDict, Field
 
-from ...consts import BrowserUseLlm
-from ...session import CreateSessionParams
+from ..consts import BrowserUseLlm
+from ..session import CreateSessionParams
 
 BrowserUseTaskStatus = Literal["pending", "running", "completed", "failed", "stopped"]
 
@@ -182,3 +183,4 @@ class BrowserUseTaskResponse(BaseModel):
     data: Optional[BrowserUseTaskData] = Field(default=None, alias="data")
     error: Optional[str] = Field(default=None, alias="error")
     live_url: Optional[str] = Field(default=None, alias="liveUrl")
+    live_url: Optional[str] = Field(default=None, alias="liveUrl")

{hyperbrowser-0.33.0 → hyperbrowser-0.34.0}/hyperbrowser/models/crawl.py

@@ -47,6 +47,18 @@ class StartCrawlJobResponse(BaseModel):
     job_id: str = Field(alias="jobId")
 
 
+class CrawlJobStatusResponse(BaseModel):
+    """
+    Response from getting the status of a crawl job.
+    """
+
+    model_config = ConfigDict(
+        populate_by_alias=True,
+    )
+
+    status: CrawlJobStatus
+
+
 class CrawledPage(BaseModel):
     """
     Data from a crawled page.

{hyperbrowser-0.33.0 → hyperbrowser-0.34.0}/hyperbrowser/models/extract.py

@@ -43,6 +43,18 @@ class StartExtractJobResponse(BaseModel):
     job_id: str = Field(alias="jobId")
 
 
+class ExtractJobStatusResponse(BaseModel):
+    """
+    Response from getting the status of a extract job.
+    """
+
+    model_config = ConfigDict(
+        populate_by_alias=True,
+    )
+
+    status: ExtractJobStatus
+
+
 class ExtractJobResponse(BaseModel):
     """
     Response from a extract job.

{hyperbrowser-0.33.0 → hyperbrowser-0.34.0}/hyperbrowser/models/scrape.py

@@ -78,6 +78,18 @@ class StartScrapeJobResponse(BaseModel):
     job_id: str = Field(alias="jobId")
 
 
+class ScrapeJobStatusResponse(BaseModel):
+    """
+    Response from getting the status of a scrape job.
+    """
+
+    model_config = ConfigDict(
+        populate_by_alias=True,
+    )
+
+    status: ScrapeJobStatus
+
+
 class ScrapeJobData(BaseModel):
     """
     Data from a scraped site.
@@ -119,6 +131,18 @@ class StartBatchScrapeJobParams(BaseModel):
     )
 
 
+class BatchScrapeJobStatusResponse(BaseModel):
+    """
+    Response from getting the status of a batch scrape job.
+    """
+
+    model_config = ConfigDict(
+        populate_by_alias=True,
+    )
+
+    status: ScrapeJobStatus
+
+
 class ScrapedPage(BaseModel):
     """
     A scraped page.

{hyperbrowser-0.33.0 → hyperbrowser-0.34.0}/hyperbrowser/tools/__init__.py

@@ -1,15 +1,18 @@
 import json
+from hyperbrowser.models.beta.agents.browser_use import StartBrowserUseTaskParams
 from hyperbrowser.models.crawl import StartCrawlJobParams
 from hyperbrowser.models.extract import StartExtractJobParams
 from hyperbrowser.models.scrape import StartScrapeJobParams
 from hyperbrowser import Hyperbrowser, AsyncHyperbrowser
 
 from .openai import (
+    BROWSER_USE_TOOL_OPENAI,
     EXTRACT_TOOL_OPENAI,
     SCRAPE_TOOL_OPENAI,
     CRAWL_TOOL_OPENAI,
 )
 from .anthropic import (
+    BROWSER_USE_TOOL_ANTHROPIC,
     EXTRACT_TOOL_ANTHROPIC,
     SCRAPE_TOOL_ANTHROPIC,
     CRAWL_TOOL_ANTHROPIC,
@@ -79,8 +82,28 @@ class WebsiteExtractTool:
         return json.dumps(resp.data) if resp.data else ""
 
 
+class BrowserUseTool:
+    openai_tool_definition = BROWSER_USE_TOOL_OPENAI
+    anthropic_tool_definition = BROWSER_USE_TOOL_ANTHROPIC
+
+    @staticmethod
+    def runnable(hb: Hyperbrowser, params: dict) -> str:
+        resp = hb.beta.agents.browser_use.start_and_wait(
+            params=StartBrowserUseTaskParams(**params)
+        )
+        return resp.data.final_result if resp.data and resp.data.final_result else ""
+
+    @staticmethod
+    async def async_runnable(hb: AsyncHyperbrowser, params: dict) -> str:
+        resp = await hb.beta.agents.browser_use.start_and_wait(
+            params=StartBrowserUseTaskParams(**params)
+        )
+        return resp.data.final_result if resp.data and resp.data.final_result else ""
+
+
 __all__ = [
     "WebsiteScrapeTool",
     "WebsiteCrawlTool",
     "WebsiteExtractTool",
+    "BrowserUseTool",
 ]

{hyperbrowser-0.33.0 → hyperbrowser-0.34.0}/hyperbrowser/tools/anthropic.py

@@ -1,7 +1,12 @@
 from typing import Dict, Union, Optional
 from typing_extensions import Literal, Required, TypeAlias, TypedDict
 
-from hyperbrowser.tools.schema import CRAWL_SCHEMA, EXTRACT_SCHEMA, SCRAPE_SCHEMA
+from hyperbrowser.tools.schema import (
+    BROWSER_USE_SCHEMA,
+    CRAWL_SCHEMA,
+    EXTRACT_SCHEMA,
+    SCRAPE_SCHEMA,
+)
 
 
 class CacheControlEphemeralParam(TypedDict, total=False):
@@ -60,3 +65,9 @@ EXTRACT_TOOL_ANTHROPIC: ToolParam = {
     "name": "extract_data",
     "description": "Extract data in a structured format from multiple URLs in a single function call. IMPORTANT: When information must be gathered from multiple sources (such as comparing items, researching topics across sites, or answering questions that span multiple webpages), ALWAYS include all relevant URLs in ONE function call. This enables comprehensive answers with cross-referenced information. Returns data as a json string.",
 }
+
+BROWSER_USE_TOOL_ANTHROPIC: ToolParam = {
+    "input_schema": BROWSER_USE_SCHEMA,
+    "name": "browser_use",
+    "description": "Have an AI agent use a browser to perform a task on the web.",
+}

{hyperbrowser-0.33.0 → hyperbrowser-0.34.0}/hyperbrowser/tools/openai.py

@@ -1,7 +1,12 @@
 from typing import Dict, Optional
 from typing_extensions import Literal, Required, TypedDict, TypeAlias
 
-from hyperbrowser.tools.schema import CRAWL_SCHEMA, EXTRACT_SCHEMA, SCRAPE_SCHEMA
+from hyperbrowser.tools.schema import (
+    BROWSER_USE_SCHEMA,
+    CRAWL_SCHEMA,
+    EXTRACT_SCHEMA,
+    SCRAPE_SCHEMA,
+)
 
 FunctionParameters: TypeAlias = Dict[str, object]
 
@@ -77,3 +82,13 @@ EXTRACT_TOOL_OPENAI: ChatCompletionToolParam = {
         "strict": True,
     },
 }
+
+BROWSER_USE_TOOL_OPENAI: ChatCompletionToolParam = {
+    "type": "function",
+    "function": {
+        "name": "browser_use",
+        "description": "Have an AI agent use a browser to perform a task on the web.",
+        "parameters": BROWSER_USE_SCHEMA,
+        "strict": True,
+    },
+}

{hyperbrowser-0.33.0 → hyperbrowser-0.34.0}/hyperbrowser/tools/schema.py

@@ -111,3 +111,50 @@ EXTRACT_SCHEMA = {
     "required": ["urls", "prompt", "schema", "max_links"],
     "additionalProperties": False,
 }
+
+BROWSER_USE_LLM_SCHEMA = {
+    "type": "string",
+    "enum": [
+        "gpt-4o",
+        "gpt-4o-mini",
+        "claude-3-7-sonnet-20250219",
+        "claude-3-5-sonnet-20241022",
+        "claude-3-5-haiku-20241022",
+        "gemini-2.0-flash",
+    ],
+    "default": "gemini-2.0-flash",
+}
+
+BROWSER_USE_SCHEMA = {
+    "type": "object",
+    "properties": {
+        "task": {
+            "type": "string",
+            "description": "The text description of the task to be performed by the agent.",
+        },
+        "llm": {
+            **BROWSER_USE_LLM_SCHEMA,
+            "description": "The language model (LLM) instance to use for generating actions. Default to gemini-2.0-flash.",
+        },
+        "planner_llm": {
+            **BROWSER_USE_LLM_SCHEMA,
+            "description": "The language model to use specifically for planning future actions, can differ from the main LLM. Default to gemini-2.0-flash.",
+        },
+        "page_extraction_llm": {
+            **BROWSER_USE_LLM_SCHEMA,
+            "description": "The language model to use for extracting structured data from webpages. Default to gemini-2.0-flash.",
+        },
+        "keep_browser_open": {
+            "type": "boolean",
+            "description": "When enabled, keeps the browser session open after task completion.",
+        },
+    },
+    "required": [
+        "task",
+        "llm",
+        "planner_llm",
+        "page_extraction_llm",
+        "keep_browser_open",
+    ],
+    "additionalProperties": False,
+}

{hyperbrowser-0.33.0 → hyperbrowser-0.34.0}/pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "hyperbrowser"
-version = "0.33.0"
+version = "0.34.0"
 description = "Python SDK for hyperbrowser"
 authors = ["Nikhil Shahi <nshahi1998@gmail.com>"]
 license = "MIT"
@@ -13,6 +13,7 @@ repository = "https://github.com/hyperbrowserai/python-sdk"
 python = "^3.8"
 pydantic = ">=2.0,<3"
 httpx = ">=0.23.0,<1"
+jsonref = ">=1.1.0"
 
 
 [tool.poetry.group.dev.dependencies]

hyperbrowser-0.33.0/hyperbrowser/client/managers/async_manager/beta/__init__.py

@@ -1,6 +0,0 @@
-from .agents import Agents
-
-
-class Beta:
-    def __init__(self, client):
-        self.agents = Agents(client)

hyperbrowser-0.33.0/hyperbrowser/client/managers/sync_manager/beta/__init__.py

@@ -1,6 +0,0 @@
-from .agents import Agents
-
-
-class Beta:
-    def __init__(self, client):
-        self.agents = Agents(client)