The agent engineering platform.
Python agentsaiai-agentsanthropicchatgptdeepagentsenterpriseframeworkgeminigenerative-ailangchainlanggraphllmmultiagentopen-sourceopenaipydanticpythonragtypescript
community: Add parameter `allow_dangerous_requests` to `WebResearchRetriever.from_llm` construct (#24712)
Cong Peng190988d93ece
diff --git a/libs/community/langchain_community/retrievers/web_research.py b/libs/community/langchain_community/retrievers/web_research.pyindex 2187225f38..8df4801392 100644--- a/libs/community/langchain_community/retrievers/web_research.py+++ b/libs/community/langchain_community/retrievers/web_research.py@@ -122,6 +122,7 @@ class WebResearchRetriever(BaseRetriever):chunk_size=1500, chunk_overlap=150),trust_env: bool = False,+ allow_dangerous_requests: bool = False,) -> "WebResearchRetriever":"""Initialize from llm using default template.@@ -162,6 +163,7 @@ class WebResearchRetriever(BaseRetriever):num_search_results=num_search_results,text_splitter=text_splitter,trust_env=trust_env,+ allow_dangerous_requests=allow_dangerous_requests,)def clean_search_query(self, query: str) -> str: