[GH-ISSUE #16558] issue: The lack of transmission of headers in rag prompt #56624

New Issue

GiteaMirror · 2026-05-05T19:49:42-05:00

GiteaMirror commented

2026-05-05 19:49:42 -05:00

Originally created by @Sank-WoT on GitHub (Aug 13, 2025).
Original GitHub issue: https://github.com/open-webui/open-webui/issues/16558

Check Existing Issues

I have searched the existing issues and discussions.
I am using the latest version of Open WebUI.

Installation Method

Git Clone

Open WebUI Version

0.6.23

Ollama Version (if applicable)

No response

Operating System

win10

Browser (if applicable)

No response

Confirmation

I have read and followed all instructions in README.md.
I am using the latest version of both Open WebUI and Ollama.
I have included the browser console logs.
I have included the Docker container logs.
I have provided every relevant configuration, setting, and environment variable used in my setup.
I have clearly listed every relevant configuration, custom setting, environment variable, and command-line option that influences my setup (such as Docker Compose overrides, .env values, browser settings, authentication configurations, etc).
I have documented step-by-step reproduction instructions that are precise, sequential, and leave nothing to interpretation. My steps:
Start with the initial platform/version/OS and dependencies used,
Specify exact install/launch/configure commands,
List URLs visited, user input (incl. example values/emails/passwords if needed),
Describe all options and toggles enabled or changed,
Include any files or environmental changes,
Identify the expected and actual result at each stage,
Ensure any reasonably skilled user can follow and hit the same issue.

Expected Behavior

Based on the description of the 0.6.16 release, it is expected that the markdown hierarchy will be preserved and passed as a prompt.

Enhanced Markdown Document Splitting: Improve knowledge retrieval from Markdown files with a new header-aware splitting strategy. This method intelligently chunks documents based on their header structure, preserving the original context and hierarchy for more accurate and relevant RAG results.

Example: I used our README.md

I expect this behavior while maintaining the header hierarchy for the docker question.

Open WebUI

How to Install

Quick Start with Docker 🐳

Note

Please note that for certain Docker environments, additional configurations might be needed. If you encounter any connection issues, our detailed guide on Open WebUI Documentation is ready to assist you.

Warning

When using Docker to install Open WebUI, make sure to include the -v open-webui:/app/backend/data in your Docker command. This step is crucial as it ensures your database is properly mounted and prevents any loss of data.

Tip

If you wish to utilize Open WebUI with Ollama included or CUDA acceleration, we recommend utilizing our official images tagged with either :cuda or :ollama. To enable CUDA, you must install the Nvidia CUDA container toolkit on your Linux/WSL system.

Actual Behavior

I got chunk

Quick Start with Docker 🐳

Note

Please note that for certain Docker environments, additional configurations might be needed. If you encounter any connection issues, our detailed guide on Open WebUI Documentation is ready to assist you.

Warning

When using Docker to install Open WebUI, make sure to include the -v open-webui:/app/backend/data in your Docker command. This step is crucial as it ensures your database is properly mounted and prevents any loss of data.

Tip

If you wish to utilize Open WebUI with Ollama included or CUDA acceleration, we recommend utilizing our official images tagged with either :cuda or :ollama. To enable CUDA, you must install the Nvidia CUDA container toolkit on your Linux/WSL system.

Steps to Reproduce

Uploaded README.md , 2. asked a question about docker

Logs & Screenshots

After examining the code, I noticed that the headers are not transmitted anywhere in the prompt.

Here the headers are collected in the metadata field
438e5d966f/backend/open_webui/routers/retrieval.py (L1220)

Here we form the context
438e5d966f/backend/open_webui/utils/middleware.py (L967)

here we form final rag template
438e5d966f/backend/open_webui/utils/task.py (L151)

After reviewing the code, I don't see highlighted headers being used anywhere.
I think this is a bug, otherwise add documentation where the headers are used?

Additional Information

No response

Originally created by @Sank-WoT on GitHub (Aug 13, 2025). Original GitHub issue: https://github.com/open-webui/open-webui/issues/16558 ### Check Existing Issues - [x] I have searched the existing issues and discussions. - [x] I am using the latest version of Open WebUI. ### Installation Method Git Clone ### Open WebUI Version 0.6.23 ### Ollama Version (if applicable) _No response_ ### Operating System win10 ### Browser (if applicable) _No response_ ### Confirmation - [x] I have read and followed all instructions in `README.md`. - [x] I am using the latest version of **both** Open WebUI and Ollama. - [x] I have included the browser console logs. - [x] I have included the Docker container logs. - [x] I have **provided every relevant configuration, setting, and environment variable used in my setup.** - [x] I have clearly **listed every relevant configuration, custom setting, environment variable, and command-line option that influences my setup** (such as Docker Compose overrides, .env values, browser settings, authentication configurations, etc). - [x] I have documented **step-by-step reproduction instructions that are precise, sequential, and leave nothing to interpretation**. My steps: - Start with the initial platform/version/OS and dependencies used, - Specify exact install/launch/configure commands, - List URLs visited, user input (incl. example values/emails/passwords if needed), - Describe all options and toggles enabled or changed, - Include any files or environmental changes, - Identify the expected and actual result at each stage, - Ensure any reasonably skilled user can follow and hit the same issue. ### Expected Behavior Based on the description of the 0.6.16 release, it is expected that the markdown hierarchy will be preserved and passed as a prompt. _Enhanced Markdown Document Splitting: Improve knowledge retrieval from Markdown files with a new header-aware splitting strategy. This method intelligently chunks documents based on their header structure, preserving the original context and **hierarchy** for more accurate and relevant RAG results._ Example: I used our README.md I expect this behavior while maintaining the header hierarchy for the docker question. # Open WebUI ## How to Install ### Quick Start with Docker 🐳 > [!NOTE] > Please note that for certain Docker environments, additional configurations might be needed. If you encounter any connection issues, our detailed guide on [Open WebUI Documentation](https://docs.openwebui.com/) is ready to assist you. > [!WARNING] > When using Docker to install Open WebUI, make sure to include the `-v open-webui:/app/backend/data` in your Docker command. This step is crucial as it ensures your database is properly mounted and prevents any loss of data. > [!TIP] > If you wish to utilize Open WebUI with Ollama included or CUDA acceleration, we recommend utilizing our official images tagged with either `:cuda` or `:ollama`. To enable CUDA, you must install the [Nvidia CUDA container toolkit](https://docs.nvidia.com/dgx/nvidia-container-runtime-upgrade/) on your Linux/WSL system. ### Actual Behavior I got chunk ### Quick Start with Docker 🐳 > [!NOTE] > Please note that for certain Docker environments, additional configurations might be needed. If you encounter any connection issues, our detailed guide on [Open WebUI Documentation](https://docs.openwebui.com/) is ready to assist you. > [!WARNING] > When using Docker to install Open WebUI, make sure to include the `-v open-webui:/app/backend/data` in your Docker command. This step is crucial as it ensures your database is properly mounted and prevents any loss of data. > [!TIP] > If you wish to utilize Open WebUI with Ollama included or CUDA acceleration, we recommend utilizing our official images tagged with either `:cuda` or `:ollama`. To enable CUDA, you must install the [Nvidia CUDA container toolkit](https://docs.nvidia.com/dgx/nvidia-container-runtime-upgrade/) on your Linux/WSL system. ### Steps to Reproduce 1. Uploaded README.md , 2. asked a question about docker ### Logs & Screenshots After examining the code, I noticed that the headers are not transmitted anywhere in the prompt. Here the headers are collected in the metadata field https://github.com/open-webui/open-webui/blob/438e5d966f0f64f9ea3feab22724a5bd96a4127b/backend/open_webui/routers/retrieval.py#L1220 <img width="822" height="613" alt="Image" src="https://github.com/user-attachments/assets/e149d8a2-0879-4597-9f91-43cf7aa70e90" /> Here we form the context https://github.com/open-webui/open-webui/blob/438e5d966f0f64f9ea3feab22724a5bd96a4127b/backend/open_webui/utils/middleware.py#L967 <img width="849" height="722" alt="Image" src="https://github.com/user-attachments/assets/b00b4468-20ca-4cd7-92a0-19bddfa5ad47" /> here we form final rag template https://github.com/open-webui/open-webui/blob/438e5d966f0f64f9ea3feab22724a5bd96a4127b/backend/open_webui/utils/task.py#L151 <img width="647" height="360" alt="Image" src="https://github.com/user-attachments/assets/78e0c2a2-a57b-4cfc-a2d1-dc6d1394a3a1" /> After reviewing the code, I don't see highlighted headers being used anywhere. I think this is a bug, otherwise add documentation where the headers are used? ### Additional Information _No response_

GiteaMirror added the bug label 2026-05-05 19:49:42 -05:00

GiteaMirror closed this issue

2026-05-05 19:49:43 -05:00

Sign in to join this conversation.

Branches Tags

1 Participants

Notifications

Due Date

No due date set.

Dependencies

No dependencies set.

Reference: github-starred/open-webui#56624