lm-sys / lm-sys/FastChat

HTML injection of the web prompt: possible safety issue?

Open
#3,154 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
39.5k
Forks
4.8k
PR merge metrics
No merged PRs in 30d

Description

When writing a prompt, one can add any HTML/js code and this will be rendered on the chatbot.

I am not sure about the security implications of this.

Screenshot of the arena when asking with an image tag:

Screenshot 2024-03-14 at 16 45 40

Contributor guide

No contributing guide indexed for this repository

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start by reproducing the reported HTML and JavaScript rendering in the web prompt, then trace where prompt content is rendered in the chatbot interface. Determine whether untrusted content can execute or affect other users, and add a regression test showing that prompt markup is handled safely.

Written by the indexing model from the issue text.

Assessment

Tech stack
html, javascript
Domain
frontend, security, web-dev
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.