Enriched input signal for Claude Code - Eyeballs for non-verbal communication cues

Preflight Checklist

• [x] I have searched existing requests and this feature hasn't been requested yet
• [x] This is a single feature request (not multiple features)

Problem Statement

I often communicate with Claude in short, terse phrases. Sometimes it misunderstands intent because it doesn't perceive the emotions inherent in my communication. The text only carries part of the communication signal. Anyone who has ever ended up in an email war knows that a single face-to-face can clear up all the misunderstandings and emotional misreads. I could type more and be clearer, but I don't want to.

Proposed Solution

Give Claude eyeballs so he has a fighting chance to read these moments correctly. Read my face and connect it with my written input. Give Claude Code access to my camera. Capture two short videos per turn. First, to record the user's reaction to a Claude response. Second, to record sentiment when the user posts their response to Claude. Like a fighter plane's gun-cam footage. Now Claude can use non-verbal cues.

Alternative Solutions

_No response_

Priority

Low - Nice to have

Feature Category

Other

Use Case Example

Claude turns ambiguous, non-imperative user observations into commands, and resolves the ambiguity with recent-task momentum biased toward action. Example: mid edit-session - prior turns had been removing "mechanism leak" phrases from a document - the user wrote a flat, non-imperative statement meant to share a eureka moment with Claude. The excitement wasn't discernible from the text alone.

Claude read the statement as "remove the phrase," deleted it, and justified the deletion because the session had established a cut-the-leaks pattern. The phrase was the actual payoff; the user had to restore it. A visual signal carrying the non-verbal communication might have smoothed this interaction.

Additional Context

_No response_