Native pdf reader like claude code in codex would be really helpful. currently it converts pdf to text and loses context in figures in pdf

[Subramaniam-dot]

Hi team could you add a native, figure-aware PDF reader mode (similar to Claude Code)?
Right now PDFs get converted to plain text and we lose figures/tables/diagrams + their captions and layout context, which breaks understanding for technical papers and reports.

[etraut-openai]

We're tracking that feature request here. Please upvote if you want to see this added. We generally prioritize our features based on upvotes from the community.