klbr/khoj
1
0
Fork 0
mirror of https://github.com/khoaliber/khoj.git synced 2026-03-10 13:26:13 +00:00
Code Issues Packages Projects Releases Wiki Activity

Disable web search tool if no search engine configured by admin

Browse Source 
Webpage read is gated behind having a web search engine configured for
now. It can later be decoupled from web search and depend on whether
any web scrapers is configured.
This commit is contained in:
Debanjum
2025-08-30 00:22:26 -07:00
parent 3aa6f8ba1f
commit 3441783d5b
3 changed files with 19 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
src/khoj/routers/helpers.py
View File

@@ -121,6 +121,7 @@ from khoj.utils.helpers import (
is_none_or_empty, is_none_or_empty,
is_operator_enabled, is_operator_enabled,
is_valid_url, is_valid_url,
is_web_search_enabled,
log_telemetry, log_telemetry,
mode_descriptions_for_llm, mode_descriptions_for_llm,
timer, timer,
@@ -367,6 +368,8 @@ async def aget_data_sources_and_output_format(
continue continue
if source == ConversationCommand.Operator and not is_operator_enabled(): if source == ConversationCommand.Operator and not is_operator_enabled():
continue continue
if source in [ConversationCommand.Online, ConversationCommand.Webpage] and not is_web_search_enabled():
continue
source_options[source.value] = description source_options[source.value] = description
if len(agent_sources) == 0 or source.value in agent_sources: if len(agent_sources) == 0 or source.value in agent_sources:
source_options_str += f'- "{source.value}": "{description}"\n' source_options_str += f'- "{source.value}": "{description}"\n'

5
src/khoj/routers/research.py
View File

@@ -36,6 +36,7 @@ from khoj.utils.helpers import (
dict_to_tuple, dict_to_tuple,
is_none_or_empty, is_none_or_empty,
is_operator_enabled, is_operator_enabled,
is_web_search_enabled,
timer, timer,
tools_for_research_llm, tools_for_research_llm,
truncate_code_context, truncate_code_context,
@@ -96,6 +97,7 @@ async def apick_next_tool(
ConversationCommand.ViewFile, ConversationCommand.ViewFile,
ConversationCommand.ListFiles, ConversationCommand.ListFiles,
] ]
web_research_tools = [ConversationCommand.SearchWeb, ConversationCommand.ReadWebpage]
input_tools_to_research_tools = { input_tools_to_research_tools = {
ConversationCommand.Notes.value: [tool.value for tool in document_research_tools], ConversationCommand.Notes.value: [tool.value for tool in document_research_tools],
ConversationCommand.Webpage.value: [ConversationCommand.ReadWebpage.value], ConversationCommand.Webpage.value: [ConversationCommand.ReadWebpage.value],
@@ -115,6 +117,9 @@ async def apick_next_tool(
# Skip showing document related tools if user has no documents # Skip showing document related tools if user has no documents
if tool in document_research_tools and not user_has_entries: if tool in document_research_tools and not user_has_entries:
continue continue
# Skip showing web search tool if agent has no access to internet
if tool in web_research_tools and not is_web_search_enabled():
continue
if tool == ConversationCommand.SemanticSearchFiles: if tool == ConversationCommand.SemanticSearchFiles:
description = tool_data.description.format(max_search_queries=max_document_searches) description = tool_data.description.format(max_search_queries=max_document_searches)
elif tool == ConversationCommand.ReadWebpage: elif tool == ConversationCommand.ReadWebpage:

11
src/khoj/utils/helpers.py
View File

@@ -794,6 +794,17 @@ def is_internet_connected():
return False return False
def is_web_search_enabled():
"""
Check if web search tool is enabled.
Set API key via env var for a supported search engine to enable it.
"""
return any(
not is_none_or_empty(os.getenv(api_key))
for api_key in ["GOOGLE_SEARCH_API_KEY", "SERPER_DEV_API_KEY", "JINA_API_KEY", "FIRECRAWL_API_KEY"]
)
def is_internal_url(url: str) -> bool: def is_internal_url(url: str) -> bool:
""" """
Check if a URL is likely to be internal/non-public. Check if a URL is likely to be internal/non-public.