[PR #1370] [MERGED] docs: add a page on creating files and showing them in chat #1337

New issue

Closed

opened 2026-10-18 14:18:57 +00:00 by bsakaguchi · 0 comments

bsakaguchi commented

2026-10-18 14:18:57 +00:00

Owner

📋 Pull Request Information

Original PR: https://github.com/open-webui/docs/pull/1370
Author: @silentoplayz
Created: 9/1/2026
Status: ✅ Merged
Merged: 9/2/2026
Merged by: @Classic298

Base: main ← Head: docs/open-terminal-file-generation

📝 Commits (1)

db6ff5c docs: add a page on creating files and showing them in chat

📊 Changes

8 files changed (+103 additions, -0 deletions)

View changed files

➕ docs/features/open-terminal/use-cases/creating-files.md (+103 -0)
➕ static/images/open-terminal-create-docx.png (+0 -0)
➕ static/images/open-terminal-create-markdown.png (+0 -0)
➕ static/images/open-terminal-create-pdf.png (+0 -0)
➕ static/images/open-terminal-create-pptx.png (+0 -0)
➕ static/images/open-terminal-create-python.png (+0 -0)
➕ static/images/open-terminal-create-svg.png (+0 -0)
➕ static/images/open-terminal-inline-download-button.png (+0 -0)

📄 Description

Summary

Open Terminal can write documents as well as read them, and the chat can put the result in front of you with a download button on it. That was not documented anywhere, and @Classic298 raised it in review on #1369 as a common feature request worth highlighting.

This adds use-cases/creating-files.md, a sibling to the existing "Analyze Documents & Data" page, covering:

what the default image can produce (docx, pdf, pptx, xlsx, svg, png charts, code and text)
generating a Word document, a PDF, a PowerPoint deck and an SVG chart, each displayed inline in the chat
where the download button is, and that the file is in the workspace either way

No linked issue. Raised in review on #1369, split out so that PR stays a screenshot refresh.

Checklist

I have reviewed the relevant documentation and matched the existing style.
This PR meets Open WebUI's contribution standards: it is accurate, relevant to users, narrowly scoped, maintainable, and not promotional content, advertising, lead generation, SEO placement, or a request to list a product, service, provider, integration, gateway, tool, or company primarily for visibility.
I understand that PRs that do not meet these standards may be closed without review and will not be merged. Repeated, low-quality, off-topic, promotional, or intentionally misleading submissions may result in the contributor being blocked from future participation in Open WebUI repositories.

Notes for reviewers

Every file in the screenshots was generated by the model during capture, then checked on disk. That check is not a formality: on the first pass the deck was described as written and did not exist, so that frame was discarded and the scene re-run. What shipped was verified as follows.

File	Check
`weekly-operations.docx`	opens with `python-docx`, 17 package parts, has a heading and a real table
`weekly-operations.pdf`	valid `%PDF-1.3`, two page objects
`weekly-operations.pptx`	opens with `python-pptx`, valid zip, three slides
`site-output.svg`	`viewBox="0 0 720 360"`, 15 text elements, values match the source CSV

The library list in the page was read off the running image rather than assumed: python-docx, python-pptx, fpdf, openpyxl, Pillow and matplotlib are present, reportlab is not, so it is not mentioned.

The download button screenshot is annotated. It is a crop of the file card with a circle drawn over the download control, since the point of that section is the affordance rather than the page around it. Every other image is a full-window 3200x1766 frame, matching the rest of the Open Terminal set.

Not included. SVG editing is described in prose (vector output is text, so the model rewrites shapes rather than regenerating a picture) but not shown as its own before and after pair. Happy to add that if it is worth a frame.

_{🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.}

## 📋 Pull Request Information **Original PR:** https://github.com/open-webui/docs/pull/1370 **Author:** [@silentoplayz](https://github.com/silentoplayz) **Created:** 9/1/2026 **Status:** ✅ Merged **Merged:** 9/2/2026 **Merged by:** [@Classic298](https://github.com/Classic298) **Base:** `main` ← **Head:** `docs/open-terminal-file-generation` --- ### 📝 Commits (1) - [`db6ff5c`](https://github.com/open-webui/docs/commit/db6ff5ce7e8133db1411d7f0dd54ae9320272d80) docs: add a page on creating files and showing them in chat ### 📊 Changes **8 files changed** (+103 additions, -0 deletions) <details> <summary>View changed files</summary> ➕ `docs/features/open-terminal/use-cases/creating-files.md` (+103 -0) ➕ `static/images/open-terminal-create-docx.png` (+0 -0) ➕ `static/images/open-terminal-create-markdown.png` (+0 -0) ➕ `static/images/open-terminal-create-pdf.png` (+0 -0) ➕ `static/images/open-terminal-create-pptx.png` (+0 -0) ➕ `static/images/open-terminal-create-python.png` (+0 -0) ➕ `static/images/open-terminal-create-svg.png` (+0 -0) ➕ `static/images/open-terminal-inline-download-button.png` (+0 -0) </details> ### 📄 Description ## Summary Open Terminal can write documents as well as read them, and the chat can put the result in front of you with a download button on it. That was not documented anywhere, and @Classic298 raised it in review on #1369 as a common feature request worth highlighting. This adds `use-cases/creating-files.md`, a sibling to the existing "Analyze Documents & Data" page, covering: - what the default image can produce (docx, pdf, pptx, xlsx, svg, png charts, code and text) - generating a Word document, a PDF, a PowerPoint deck and an SVG chart, each displayed inline in the chat - where the download button is, and that the file is in the workspace either way ## Related issue or discussion No linked issue. Raised in review on #1369, split out so that PR stays a screenshot refresh. ## Checklist - [x] I have reviewed the relevant documentation and matched the existing style. - [x] This PR meets Open WebUI's contribution standards: it is accurate, relevant to users, narrowly scoped, maintainable, and not promotional content, advertising, lead generation, SEO placement, or a request to list a product, service, provider, integration, gateway, tool, or company primarily for visibility. - [x] I understand that PRs that do not meet these standards may be closed without review and will not be merged. Repeated, low-quality, off-topic, promotional, or intentionally misleading submissions may result in the contributor being blocked from future participation in Open WebUI repositories. ## Notes for reviewers **Every file in the screenshots was generated by the model during capture, then checked on disk.** That check is not a formality: on the first pass the deck was described as written and did not exist, so that frame was discarded and the scene re-run. What shipped was verified as follows. | File | Check | | :--- | :--- | | `weekly-operations.docx` | opens with `python-docx`, 17 package parts, has a heading and a real table | | `weekly-operations.pdf` | valid `%PDF-1.3`, two page objects | | `weekly-operations.pptx` | opens with `python-pptx`, valid zip, three slides | | `site-output.svg` | `viewBox="0 0 720 360"`, 15 text elements, values match the source CSV | The library list in the page was read off the running image rather than assumed: `python-docx`, `python-pptx`, `fpdf`, `openpyxl`, `Pillow` and `matplotlib` are present, `reportlab` is not, so it is not mentioned. **The download button screenshot is annotated.** It is a crop of the file card with a circle drawn over the download control, since the point of that section is the affordance rather than the page around it. Every other image is a full-window 3200x1766 frame, matching the rest of the Open Terminal set. **Not included.** SVG editing is described in prose (vector output is text, so the model rewrites shapes rather than regenerating a picture) but not shown as its own before and after pair. Happy to add that if it is worth a frame. --- <sub>🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.</sub>

bsakaguchi

2026-10-18 14:18:57 +00:00