vxcontrol/pentagi
1
0
Fork 0
mirror of https://github.com/vxcontrol/pentagi.git synced 2026-07-20 12:37:04 -04:00
Code Issues 45 Packages Projects Releases 9 Wiki Activity

[GH-ISSUE #73] [Enhancement]: Azure OpenAI LLM provider #41

New Issue
Closed
opened 2026-06-06 22:08:12 -04:00 by yindo · 13 comments
yindo commented 2026-06-06 22:08:12 -04:00
Owner
Copy Link
Copy Source

Originally created by @adriaanvermaak on GitHub (Oct 28, 2025).
Original GitHub issue: https://github.com/vxcontrol/pentagi/issues/73

Originally assigned to: @asdek on GitHub.

Target Component

External Integrations (LLM/Search APIs)

Enhancement Description

good Day team,

Please could you look into adding Azure OpenAi support or indicate how to add azureAi to the custom configs?

AzureAi has lots of reasoning models available to deploy like gpt4-mini etc.

thank you for your great work

Technical Details

No response

Designs and Mockups

No response

Alternative Solutions

No response

Verification

  • I have checked that this enhancement hasn't been already proposed
  • This enhancement aligns with PentAGI's goal of autonomous penetration testing
  • I have considered the security implications of this enhancement
  • I have provided clear use cases and benefits
Originally created by @adriaanvermaak on GitHub (Oct 28, 2025). Original GitHub issue: https://github.com/vxcontrol/pentagi/issues/73 Originally assigned to: @asdek on GitHub. ### Target Component External Integrations (LLM/Search APIs) ### Enhancement Description good Day team, Please could you look into adding Azure OpenAi support or indicate how to add azureAi to the custom configs? AzureAi has lots of reasoning models available to deploy like gpt4-mini etc. thank you for your great work ### Technical Details _No response_ ### Designs and Mockups _No response_ ### Alternative Solutions _No response_ ### Verification - [ ] I have checked that this enhancement hasn't been already proposed - [ ] This enhancement aligns with PentAGI's goal of autonomous penetration testing - [ ] I have considered the security implications of this enhancement - [ ] I have provided clear use cases and benefits
yindo added the enhancement label 2026-06-06 22:08:12 -04:00
yindo closed this issue 2026-06-06 22:08:12 -04:00
yindo commented 2026-06-06 22:08:13 -04:00
Author
Owner
Copy Link
Copy Source

@asdek commented on GitHub (Oct 28, 2025):

Hello @adriaanvermaak,

We have this feature in our roadmap for the nearest future. In the meantime, the easiest way to integrate Azure OpenAI is to set up LiteLLM as a proxy and add your Azure models like gpt-5-mini, gpt-4.1, gpt-4.1-mini, o4-mini and gpt-5, if it's available in your region.

Then you can use the following environment variables:

LLM_SERVER_URL=http://litellm:4000/
LLM_SERVER_KEY=your_litellm_api_key
LLM_SERVER_CONFIG_PATH=/opt/pentagi/conf/your_custom_litellm.provider.yml
LLM_SERVER_LEGACY_REASONING=true

If you're running LiteLLM on the same server, you can place the container directly into the pentagi-network Docker network and use its hostname directly in the LLM_SERVER_URL variable.

For the LiteLLM configuration, you can use our existing OpenAI config as a template: https://github.com/vxcontrol/pentagi/blob/master/examples/configs/custom-openai.provider.yml

Alternatively, you can configure this directly from the product interface on the Settings page.

<!-- gh-comment-id:3458879184 --> @asdek commented on GitHub (Oct 28, 2025): Hello @adriaanvermaak, We have this feature in our roadmap for the nearest future. In the meantime, the easiest way to integrate Azure OpenAI is to set up LiteLLM as a proxy and add your Azure models like gpt-5-mini, gpt-4.1, gpt-4.1-mini, o4-mini and gpt-5, if it's available in your region. Then you can use the following environment variables: ``` LLM_SERVER_URL=http://litellm:4000/ LLM_SERVER_KEY=your_litellm_api_key LLM_SERVER_CONFIG_PATH=/opt/pentagi/conf/your_custom_litellm.provider.yml LLM_SERVER_LEGACY_REASONING=true ``` If you're running LiteLLM on the same server, you can place the container directly into the `pentagi-network` Docker network and use its hostname directly in the `LLM_SERVER_URL` variable. For the LiteLLM configuration, you can use our existing OpenAI config as a template: https://github.com/vxcontrol/pentagi/blob/master/examples/configs/custom-openai.provider.yml Alternatively, you can configure this directly from the product interface on the Settings page.
yindo commented 2026-06-06 22:08:13 -04:00
Author
Owner
Copy Link
Copy Source

@adriaanvermaak commented on GitHub (Dec 4, 2025):

Hi Team,

hope you are well.

I have managed to get the LiteLLM routing to take place via Azure and works perfectly.

However I would like to know how do I get the embedding model hosted via Azure to work as well?

I see in the env there are variables for cust embeddings.

Could you provide assistance in how to get it work?

Embedding

EMBEDDING_URL=
EMBEDDING_KEY=
EMBEDDING_MODEL=
EMBEDDING_PROVIDER=
EMBEDDING_BATCH_SIZE=
EMBEDDING_STRIP_NEW_LINES=

below logs for reference.

time="2025-12-04T14:54:10Z" level=error msg="failed to perform subtasks generator" error="failed to get subtasks generator result: failed to exec function 'memorist': reached max retries to call function: failed to execute handler: failed to get memorist result: failed to get task memorist result: failed to exec function 'search_in_memory': reached max retries to call function: failed to execute handler: failed to search for similar documents: error embedding query: failed to create openai embeddings: send request: Post "https://api.openai.com/v1/embeddings": dial tcp: lookup api.openai.com on 127.0.0.11:53: read udp 127.0.0.1:37387->127.0.0.11:53: i/o timeout"

time="2025-12-04T14:54:10Z" level=error msg="failed to process input" component=worker error="failed to create task for flow 45: failed to create task worker: failed to generate subtasks: failed to generate subtasks for task 18: failed to perform subtasks generator: failed to get subtasks generator result: failed to exec function 'memorist': reached max retries to call function: failed to execute handler: failed to get memorist result: failed to get task memorist result: failed to exec function 'search_in_memory': reached max retries to call function: failed to execute handler: failed to search for similar documents: error embedding query: failed to create openai embeddings: send request: Post "https://api.openai.com/v1/embeddings": dial tcp: lookup api.openai.com on 127.0.0.11:53: read udp 127.0.0.1:37387->127.0.0.11:53: i/o timeout" flow_id=45 input="?" trace_id=3157124e-17a9-4761-bc02-0a98a4459f55 user_id=1

<!-- gh-comment-id:3612686463 --> @adriaanvermaak commented on GitHub (Dec 4, 2025): Hi Team, hope you are well. I have managed to get the LiteLLM routing to take place via Azure and works perfectly. However I would like to know how do I get the embedding model hosted via Azure to work as well? I see in the env there are variables for cust embeddings. Could you provide assistance in how to get it work? ## Embedding EMBEDDING_URL= EMBEDDING_KEY= EMBEDDING_MODEL= EMBEDDING_PROVIDER= EMBEDDING_BATCH_SIZE= EMBEDDING_STRIP_NEW_LINES= below logs for reference. time="2025-12-04T14:54:10Z" level=error msg="failed to perform subtasks generator" error="failed to get subtasks generator result: failed to exec function 'memorist': reached max retries to call function: failed to execute handler: failed to get memorist result: failed to get task memorist result: failed to exec function 'search_in_memory': reached max retries to call function: failed to execute handler: failed to search for similar documents: error embedding query: failed to create openai embeddings: send request: Post \"https://api.openai.com/v1/embeddings\": dial tcp: lookup api.openai.com on 127.0.0.11:53: read udp 127.0.0.1:37387->127.0.0.11:53: i/o timeout" time="2025-12-04T14:54:10Z" level=error msg="failed to process input" component=worker error="failed to create task for flow 45: failed to create task worker: failed to generate subtasks: failed to generate subtasks for task 18: failed to perform subtasks generator: failed to get subtasks generator result: failed to exec function 'memorist': reached max retries to call function: failed to execute handler: failed to get memorist result: failed to get task memorist result: failed to exec function 'search_in_memory': reached max retries to call function: failed to execute handler: failed to search for similar documents: error embedding query: failed to create openai embeddings: send request: Post \"https://api.openai.com/v1/embeddings\": dial tcp: lookup api.openai.com on 127.0.0.11:53: read udp 127.0.0.1:37387->127.0.0.11:53: i/o timeout" flow_id=45 input="?" trace_id=3157124e-17a9-4761-bc02-0a98a4459f55 user_id=1
yindo commented 2026-06-06 22:08:13 -04:00
Author
Owner
Copy Link
Copy Source

@asdek commented on GitHub (Dec 4, 2025):

hey @adriaanvermaak

it's nice to hear you were able to set it up 😉

here are the settings that work with the latest LiteLLM version:

EMBEDDING_URL=https://litellm.example.com/v1
EMBEDDING_KEY=YOUR_LITELLM_KEY
EMBEDDING_MODEL=azure/text-embedding-3-large
EMBEDDING_PROVIDER=openai
EMBEDDING_BATCH_SIZE=
EMBEDDING_STRIP_NEW_LINES=
<!-- gh-comment-id:3612840705 --> @asdek commented on GitHub (Dec 4, 2025): hey @adriaanvermaak it's nice to hear you were able to set it up 😉 here are the settings that work with the latest LiteLLM version: ``` EMBEDDING_URL=https://litellm.example.com/v1 EMBEDDING_KEY=YOUR_LITELLM_KEY EMBEDDING_MODEL=azure/text-embedding-3-large EMBEDDING_PROVIDER=openai EMBEDDING_BATCH_SIZE= EMBEDDING_STRIP_NEW_LINES= ```
yindo commented 2026-06-06 22:08:13 -04:00
Author
Owner
Copy Link
Copy Source

@adriaanvermaak commented on GitHub (Dec 4, 2025):

hi @asdek

thank you for the quick response.

after adding the env and restarting the container

Embedding

EMBEDDING_URL=http://litellm:4000/v1
EMBEDDING_KEY=llm_master_******
EMBEDDING_MODEL=azure/aim-text-embedding-3-large-lab
EMBEDDING_PROVIDER=openai
EMBEDDING_BATCH_SIZE=
EMBEDDING_STRIP_NEW_LINES=

see this in the logs

time="2025-12-04T17:10:01Z" level=warning msg="failed to exec function" agent=custom error="failed to execute handler: failed to search for similar documents: error embedding query: failed to create openai embeddings: send request: Post "https://api.openai.com/v1/embeddings": dial tcp: lookup api.openai.com on 127.0.0.11:53: read udp 127.0.0.1:44609->127.0.0.11:53: i/o timeout" flow_id=45 func_args="{"question": "What task is currently in progress and what is the objective for the user?", "task_id": 18, "message": "Clarifying the current task and its objective."}" func_name=search_in_memory msg_chain_id=785 tool_call_id=call_TiAasoIvmZQRcXPfMLdx2Gzq

time="2025-12-04T17:10:12Z" level=error msg="failed to search for similar documents" args="{"question":"What task is currently in progress and what is the objective for the user?","task_id":18,"message":"Clarifying the current task and its objective."}" error="error embedding query: failed to create openai embeddings: send request: Post "https://api.openai.com/v1/embeddings": dial tcp: lookup api.openai.com on 127.0.0.11:53: read udp 127.0.0.1:41971->127.0.0.11:53: i/o timeout" query="What task is currently in progress and what is the objective for the user?" task_id=18 tool=search_in_memory

time="2025-12-04T17:10:12Z" level=warning msg="failed to exec function" agent=custom error="failed to execute handler: failed to search for similar documents: error embedding query: failed to create openai embeddings: send request: Post "https://api.openai.com/v1/embeddings": dial tcp: lookup api.openai.com on 127.0.0.11:53: read udp 127.0.0.1:41971->127.0.0.11:53: i/o timeout" flow_id=45 func_args="{"question": "What task is currently in progress and what is the objective for the user?", "task_id": 18, "message": "Clarifying the current task and its objective."}" func_name=search_in_memory msg_chain_id=785 tool_call_id=call_TiAasoIvmZQRcXPfMLdx2Gzq

time="2025-12-04T17:10:13Z" level=error msg="failed to exec function" agent=custom error="reached max retries to call function: failed to execute handler: failed to search for similar documents: error embedding query: failed to create openai embeddings: send request: Post "https://api.openai.com/v1/embeddings": dial tcp: lookup api.openai.com on 127.0.0.11:53: read udp 127.0.0.1:41971->127.0.0.11:53: i/o timeout" flow_id=45 func_args="{"question": "What task is currently in progress and what is the objective for the user?", "task_id": 18, "message": "Clarifying the current task and its objective."}" func_name=search_in_memory msg_chain_id=785 tool_call_id=call_TiAasoIvmZQRcXPfMLdx2Gzq

time="2025-12-04T17:10:13Z" level=error msg="failed to exec tool call" agent=searcher error="failed to exec function 'search_in_memory': reached max retries to call function: failed to execute handler: failed to search for similar documents: error embedding query: failed to create openai embeddings: send request: Post "https://api.openai.com/v1/embeddings": dial tcp: lookup api.openai.com on 127.0.0.11:53: read udp 127.0.0.1:41971->127.0.0.11:53: i/o timeout" flow_id=45 func_args="{"question": "What task is currently in progress and what is the objective for the user?", "task_id": 18, "message": "Clarifying the current task and its objective."}" func_name=search_in_memory msg_chain_id=785 provider=custom task_id=19

time="2025-12-04T17:10:13Z" level=error msg="failed to get memorist result" error="failed to get task memorist result: failed to exec function 'search_in_memory': reached max retries to call function: failed to execute handler: failed to search for similar documents: error embedding query: failed to create openai embeddings: send request: Post "https://api.openai.com/v1/embeddings": dial tcp: lookup api.openai.com on 127.0.0.11:53: read udp 127.0.0.1:41971->127.0.0.11:53: i/o timeout"

time="2025-12-04T17:10:13Z" level=warning msg="failed to exec function" agent=custom error="failed to execute handler: failed to get memorist result: failed to get task memorist result: failed to exec function 'search_in_memory': reached max retries to call function: failed to execute handler: failed to search for similar documents: error embedding query: failed to create openai embeddings: send request: Post "https://api.openai.com/v1/embeddings": dial tcp: lookup api.openai.com on 127.0.0.11:53: read udp 127.0.0.1:41971->127.0.0.11:53: i/o timeout" flow_id=45 func_args="{"question":"I need to understand what task is currently in progress and what the objective is for the user. The previous tasks indicate a quick recon assessment on a website using curl and nmap, but there is no specific task input provided. Can you clarify the current focus?","task_id":18,"message":"Get clarity on the current user's objectives and any tasks that need to be accomplished."}" func_name=memorist msg_chain_id=784 tool_call_id=call_BCKrPm9EOblOt2ghiY9ioEBx

time="2025-12-04T17:10:30Z" level=error msg="failed to search for similar documents" args="{"question": "What task is currently in progress regarding user authentication using OAuth 2.0?", "task_id": 18, "message": "Clarifying the current task in progress related to user authentication."}" error="error embedding query: failed to create openai embeddings: send request: Post "https://api.openai.com/v1/embeddings": dial tcp: lookup api.openai.com on 127.0.0.11:53: read udp 127.0.0.1:59689->127.0.0.11:53: i/o timeout" query="What task is currently in progress regarding user authentication using OAuth 2.0?" task_id=18 tool=search_in_memory

time="2025-12-04T17:10:30Z" level=warning msg="failed to exec function" agent=custom error="failed to execute handler: failed to search for similar documents: error embedding query: failed to create openai embeddings: send request: Post "https://api.openai.com/v1/embeddings": dial tcp: lookup api.openai.com on 127.0.0.11:53: read udp 127.0.0.1:59689->127.0.0.11:53: i/o timeout" flow_id=45 func_args="{"question": "What task is currently in progress regarding user authentication using OAuth 2.0?", "task_id": 18, "message": "Clarifying the current task in progress related to user authentication."}" func_name=search_in_memory msg_chain_id=791 tool_call_id=call_zji3J7UTY1rAdRkWJV7d0IKJ

time="2025-12-04T17:10:41Z" level=error msg="failed to search for similar documents" args="{"question":"What task is currently in progress regarding user authentication using OAuth 2.0?","task_id":18,"message":"Clarifying the current task in progress related to user authentication.","subtask_id":null}" error="error embedding query: failed to create openai embeddings: send request: Post "https://api.openai.com/v1/embeddings": dial tcp: lookup api.openai.com on 127.0.0.11:53: read udp 127.0.0.1:45294->127.0.0.11:53: i/o timeout" query="What task is currently in progress regarding user authentication using OAuth 2.0?" task_id=18 tool=search_in_memory

time="2025-12-04T17:10:41Z" level=warning msg="failed to exec function" agent=custom error="failed to execute handler: failed to search for similar documents: error embedding query: failed to create openai embeddings: send request: Post "https://api.openai.com/v1/embeddings": dial tcp: lookup api.openai.com on 127.0.0.11:53: read udp 127.0.0.1:45294->127.0.0.11:53: i/o timeout" flow_id=45 func_args="{"question": "What task is currently in progress regarding user authentication using OAuth 2.0?", "task_id": 18, "message": "Clarifying the current task in progress related to user authentication."}" func_name=search_in_memory msg_chain_id=791 tool_call_id=call_zji3J7UTY1rAdRkWJV7d0IKJ

<!-- gh-comment-id:3613326516 --> @adriaanvermaak commented on GitHub (Dec 4, 2025): hi @asdek thank you for the quick response. after adding the env and restarting the container ## Embedding EMBEDDING_URL=http://litellm:4000/v1 EMBEDDING_KEY=llm_master_****** EMBEDDING_MODEL=azure/aim-text-embedding-3-large-lab EMBEDDING_PROVIDER=openai EMBEDDING_BATCH_SIZE= EMBEDDING_STRIP_NEW_LINES= see this in the logs time="2025-12-04T17:10:01Z" level=warning msg="failed to exec function" agent=custom error="failed to execute handler: failed to search for similar documents: error embedding query: failed to create openai embeddings: send request: Post \"https://api.openai.com/v1/embeddings\": dial tcp: lookup api.openai.com on 127.0.0.11:53: read udp 127.0.0.1:44609->127.0.0.11:53: i/o timeout" flow_id=45 func_args="{\"question\": \"What task is currently in progress and what is the objective for the user?\", \"task_id\": 18, \"message\": \"Clarifying the current task and its objective.\"}" func_name=search_in_memory msg_chain_id=785 tool_call_id=call_TiAasoIvmZQRcXPfMLdx2Gzq time="2025-12-04T17:10:12Z" level=error msg="failed to search for similar documents" args="{\"question\":\"What task is currently in progress and what is the objective for the user?\",\"task_id\":18,\"message\":\"Clarifying the current task and its objective.\"}" error="error embedding query: failed to create openai embeddings: send request: Post \"https://api.openai.com/v1/embeddings\": dial tcp: lookup api.openai.com on 127.0.0.11:53: read udp 127.0.0.1:41971->127.0.0.11:53: i/o timeout" query="What task is currently in progress and what is the objective for the user?" task_id=18 tool=search_in_memory time="2025-12-04T17:10:12Z" level=warning msg="failed to exec function" agent=custom error="failed to execute handler: failed to search for similar documents: error embedding query: failed to create openai embeddings: send request: Post \"https://api.openai.com/v1/embeddings\": dial tcp: lookup api.openai.com on 127.0.0.11:53: read udp 127.0.0.1:41971->127.0.0.11:53: i/o timeout" flow_id=45 func_args="{\"question\": \"What task is currently in progress and what is the objective for the user?\", \"task_id\": 18, \"message\": \"Clarifying the current task and its objective.\"}" func_name=search_in_memory msg_chain_id=785 tool_call_id=call_TiAasoIvmZQRcXPfMLdx2Gzq time="2025-12-04T17:10:13Z" level=error msg="failed to exec function" agent=custom error="reached max retries to call function: failed to execute handler: failed to search for similar documents: error embedding query: failed to create openai embeddings: send request: Post \"https://api.openai.com/v1/embeddings\": dial tcp: lookup api.openai.com on 127.0.0.11:53: read udp 127.0.0.1:41971->127.0.0.11:53: i/o timeout" flow_id=45 func_args="{\"question\": \"What task is currently in progress and what is the objective for the user?\", \"task_id\": 18, \"message\": \"Clarifying the current task and its objective.\"}" func_name=search_in_memory msg_chain_id=785 tool_call_id=call_TiAasoIvmZQRcXPfMLdx2Gzq time="2025-12-04T17:10:13Z" level=error msg="failed to exec tool call" agent=searcher error="failed to exec function 'search_in_memory': reached max retries to call function: failed to execute handler: failed to search for similar documents: error embedding query: failed to create openai embeddings: send request: Post \"https://api.openai.com/v1/embeddings\": dial tcp: lookup api.openai.com on 127.0.0.11:53: read udp 127.0.0.1:41971->127.0.0.11:53: i/o timeout" flow_id=45 func_args="{\"question\": \"What task is currently in progress and what is the objective for the user?\", \"task_id\": 18, \"message\": \"Clarifying the current task and its objective.\"}" func_name=search_in_memory msg_chain_id=785 provider=custom task_id=19 time="2025-12-04T17:10:13Z" level=error msg="failed to get memorist result" error="failed to get task memorist result: failed to exec function 'search_in_memory': reached max retries to call function: failed to execute handler: failed to search for similar documents: error embedding query: failed to create openai embeddings: send request: Post \"https://api.openai.com/v1/embeddings\": dial tcp: lookup api.openai.com on 127.0.0.11:53: read udp 127.0.0.1:41971->127.0.0.11:53: i/o timeout" time="2025-12-04T17:10:13Z" level=warning msg="failed to exec function" agent=custom error="failed to execute handler: failed to get memorist result: failed to get task memorist result: failed to exec function 'search_in_memory': reached max retries to call function: failed to execute handler: failed to search for similar documents: error embedding query: failed to create openai embeddings: send request: Post \"https://api.openai.com/v1/embeddings\": dial tcp: lookup api.openai.com on 127.0.0.11:53: read udp 127.0.0.1:41971->127.0.0.11:53: i/o timeout" flow_id=45 func_args="{\"question\":\"I need to understand what task is currently in progress and what the objective is for the user. The previous tasks indicate a quick recon assessment on a website using curl and nmap, but there is no specific task input provided. Can you clarify the current focus?\",\"task_id\":18,\"message\":\"Get clarity on the current user's objectives and any tasks that need to be accomplished.\"}" func_name=memorist msg_chain_id=784 tool_call_id=call_BCKrPm9EOblOt2ghiY9ioEBx time="2025-12-04T17:10:30Z" level=error msg="failed to search for similar documents" args="{\"question\": \"What task is currently in progress regarding user authentication using OAuth 2.0?\", \"task_id\": 18, \"message\": \"Clarifying the current task in progress related to user authentication.\"}" error="error embedding query: failed to create openai embeddings: send request: Post \"https://api.openai.com/v1/embeddings\": dial tcp: lookup api.openai.com on 127.0.0.11:53: read udp 127.0.0.1:59689->127.0.0.11:53: i/o timeout" query="What task is currently in progress regarding user authentication using OAuth 2.0?" task_id=18 tool=search_in_memory time="2025-12-04T17:10:30Z" level=warning msg="failed to exec function" agent=custom error="failed to execute handler: failed to search for similar documents: error embedding query: failed to create openai embeddings: send request: Post \"https://api.openai.com/v1/embeddings\": dial tcp: lookup api.openai.com on 127.0.0.11:53: read udp 127.0.0.1:59689->127.0.0.11:53: i/o timeout" flow_id=45 func_args="{\"question\": \"What task is currently in progress regarding user authentication using OAuth 2.0?\", \"task_id\": 18, \"message\": \"Clarifying the current task in progress related to user authentication.\"}" func_name=search_in_memory msg_chain_id=791 tool_call_id=call_zji3J7UTY1rAdRkWJV7d0IKJ time="2025-12-04T17:10:41Z" level=error msg="failed to search for similar documents" args="{\"question\":\"What task is currently in progress regarding user authentication using OAuth 2.0?\",\"task_id\":18,\"message\":\"Clarifying the current task in progress related to user authentication.\",\"subtask_id\":null}" error="error embedding query: failed to create openai embeddings: send request: Post \"https://api.openai.com/v1/embeddings\": dial tcp: lookup api.openai.com on 127.0.0.11:53: read udp 127.0.0.1:45294->127.0.0.11:53: i/o timeout" query="What task is currently in progress regarding user authentication using OAuth 2.0?" task_id=18 tool=search_in_memory time="2025-12-04T17:10:41Z" level=warning msg="failed to exec function" agent=custom error="failed to execute handler: failed to search for similar documents: error embedding query: failed to create openai embeddings: send request: Post \"https://api.openai.com/v1/embeddings\": dial tcp: lookup api.openai.com on 127.0.0.11:53: read udp 127.0.0.1:45294->127.0.0.11:53: i/o timeout" flow_id=45 func_args="{\"question\": \"What task is currently in progress regarding user authentication using OAuth 2.0?\", \"task_id\": 18, \"message\": \"Clarifying the current task in progress related to user authentication.\"}" func_name=search_in_memory msg_chain_id=791 tool_call_id=call_zji3J7UTY1rAdRkWJV7d0IKJ
yindo commented 2026-06-06 22:08:13 -04:00
Author
Owner
Copy Link
Copy Source

@adriaanvermaak commented on GitHub (Dec 4, 2025):

not sure if we need to add anything else to the custom provider yml but seems to persist with using openai

<!-- gh-comment-id:3613356414 --> @adriaanvermaak commented on GitHub (Dec 4, 2025): not sure if we need to add anything else to the custom provider yml but seems to persist with using openai
yindo commented 2026-06-06 22:08:13 -04:00
Author
Owner
Copy Link
Copy Source

@asdek commented on GitHub (Dec 4, 2025):

it's so weird... please, be sure that you make docker compose up -d in the pentagi directory, because it should use your EMBEDDING_URL options.
also, I'll check it in 2 hours.

<!-- gh-comment-id:3613362419 --> @asdek commented on GitHub (Dec 4, 2025): it's so weird... please, be sure that you make `docker compose up -d` in the pentagi directory, because it should use your EMBEDDING_URL options. also, I'll check it in 2 hours.
yindo commented 2026-06-06 22:08:14 -04:00
Author
Owner
Copy Link
Copy Source

@adriaanvermaak commented on GitHub (Dec 4, 2025):

thanks @asdek ,

will be available as well.

compose is running in correct directory for pentagi.

I have also commented out openai variable in .env as well, no difference.

happy share configs if needed.

<!-- gh-comment-id:3613508828 --> @adriaanvermaak commented on GitHub (Dec 4, 2025): thanks @asdek , will be available as well. compose is running in correct directory for pentagi. I have also commented out openai variable in .env as well, no difference. happy share configs if needed.
yindo commented 2026-06-06 22:08:14 -04:00
Author
Owner
Copy Link
Copy Source

@asdek commented on GitHub (Dec 4, 2025):

@adriaanvermaak I checked it on my side and these vars enough to work, also, by the code it should work...

EMBEDDING_URL=https://litellm.example.com/v1
EMBEDDING_KEY=sk-********
EMBEDDING_MODEL=azure/text-embedding-3-large
EMBEDDING_PROVIDER=openai

let's check your docker-compose.yml, is it contain these strings for pentagi service?

      - EMBEDDING_URL=${EMBEDDING_URL:-}
      - EMBEDDING_KEY=${EMBEDDING_KEY:-}
      - EMBEDDING_MODEL=${EMBEDDING_MODEL:-}
      - EMBEDDING_PROVIDER=${EMBEDDING_PROVIDER:-}

in additional, you can use etester tool to check embedder before run a flow:

# just for control variables in the container, you should get your values from .env file
docker exec -it pentagi env | grep EMBEDDING

# simple test, you should get "All tests passed successfully!"
docker exec -it pentagi ./bin/etester test

# any question to check errors
docker exec -it pentagi ./bin/etester search -query "What penetration testing is?"
<!-- gh-comment-id:3614159914 --> @asdek commented on GitHub (Dec 4, 2025): @adriaanvermaak I checked it on my side and these vars enough to work, also, by the code it should work... ``` EMBEDDING_URL=https://litellm.example.com/v1 EMBEDDING_KEY=sk-******** EMBEDDING_MODEL=azure/text-embedding-3-large EMBEDDING_PROVIDER=openai ``` let's check your docker-compose.yml, is it contain these strings for pentagi service? ``` - EMBEDDING_URL=${EMBEDDING_URL:-} - EMBEDDING_KEY=${EMBEDDING_KEY:-} - EMBEDDING_MODEL=${EMBEDDING_MODEL:-} - EMBEDDING_PROVIDER=${EMBEDDING_PROVIDER:-} ``` in additional, you can use `etester` tool to check embedder before run a flow: ``` # just for control variables in the container, you should get your values from .env file docker exec -it pentagi env | grep EMBEDDING # simple test, you should get "All tests passed successfully!" docker exec -it pentagi ./bin/etester test # any question to check errors docker exec -it pentagi ./bin/etester search -query "What penetration testing is?" ```
yindo commented 2026-06-06 22:08:14 -04:00
Author
Owner
Copy Link
Copy Source

@adriaanvermaak commented on GitHub (Dec 4, 2025):

thank you @asdek

seems just updating the env with those values for embedding doesnt populate into the container.

╰─ docker exec -it pentagi env | grep EMBEDDING

EMBEDDING_PROVIDER=
EMBEDDING_URL=
EMBEDDING_KEY=
EMBEDDING_MODEL=
EMBEDDING_STRIP_NEW_LINES=
EMBEDDING_BATCH_SIZE=

after docker compose up -d --force-recreate pentagi

docker exec -it pentagi env | grep EMBEDDING

EMBEDDING_KEY=llm_***********************
EMBEDDING_MODEL=azure/aim-text-embedding-3-large-lab
EMBEDDING_BATCH_SIZE=
EMBEDDING_STRIP_NEW_LINES=
EMBEDDING_PROVIDER=openai
EMBEDDING_URL=http://litellm:4000/v1

thanks @asdek appreciate your help

<!-- gh-comment-id:3614196733 --> @adriaanvermaak commented on GitHub (Dec 4, 2025): thank you @asdek seems just updating the env with those values for embedding doesnt populate into the container. ╰─ docker exec -it pentagi env | grep EMBEDDING EMBEDDING_PROVIDER= EMBEDDING_URL= EMBEDDING_KEY= EMBEDDING_MODEL= EMBEDDING_STRIP_NEW_LINES= EMBEDDING_BATCH_SIZE= after docker compose up -d --force-recreate pentagi docker exec -it pentagi env | grep EMBEDDING EMBEDDING_KEY=llm_*********************** EMBEDDING_MODEL=azure/aim-text-embedding-3-large-lab EMBEDDING_BATCH_SIZE= EMBEDDING_STRIP_NEW_LINES= EMBEDDING_PROVIDER=openai EMBEDDING_URL=http://litellm:4000/v1 thanks @asdek appreciate your help
yindo commented 2026-06-06 22:08:14 -04:00
Author
Owner
Copy Link
Copy Source

@asdek commented on GitHub (Dec 4, 2025):

you are welcome!

<!-- gh-comment-id:3614200620 --> @asdek commented on GitHub (Dec 4, 2025): you are welcome!
yindo commented 2026-06-06 22:08:14 -04:00
Author
Owner
Copy Link
Copy Source

@adriaanvermaak commented on GitHub (Dec 4, 2025):

hi @asdek

sorry for dropping this question in here.

for the kali containers that spin up with the workflows, where can we add global DNS settings for them?

<!-- gh-comment-id:3614269869 --> @adriaanvermaak commented on GitHub (Dec 4, 2025): hi @asdek sorry for dropping this question in here. for the kali containers that spin up with the workflows, where can we add global DNS settings for them?
yindo commented 2026-06-06 22:08:14 -04:00
Author
Owner
Copy Link
Copy Source

@asdek commented on GitHub (Dec 4, 2025):

I think you can use file /etc/docker/daemon.json and there should be dns-servers option.
I would recommend you to use separate installation for two nodes according to this guide https://github.com/vxcontrol/pentagi/blob/master/examples/guides/worker_node.md
And this settings about dns-servers you should do on the worker node only.

<!-- gh-comment-id:3614396925 --> @asdek commented on GitHub (Dec 4, 2025): I think you can use file /etc/docker/daemon.json and there should be dns-servers option. I would recommend you to use separate installation for two nodes according to this guide https://github.com/vxcontrol/pentagi/blob/master/examples/guides/worker_node.md And this settings about dns-servers you should do on the worker node only.
yindo commented 2026-06-06 22:08:14 -04:00
Author
Owner
Copy Link
Copy Source

@adriaanvermaak commented on GitHub (Dec 4, 2025):

Thanks you for your great assistance @asdek

<!-- gh-comment-id:3614456448 --> @adriaanvermaak commented on GitHub (Dec 4, 2025): Thanks you for your great assistance @asdek
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
bug
enhancement
help wanted
pull-request
Mirrored from GitHub Pull Request
 question
No Label enhancement
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
yindo
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: vxcontrol/pentagi#41
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?