Kiss Cam

Turns two reference photos into a convincing fake arena Kiss Cam moment, complete with a Jumbotron still and a 15-second video clip. The pipeline uses GPT-image-2 for the spectator phone shot (retro Kiss Cam graphic, scoreboard, arena lighting) then Kling v3-omni for the video with PA announcer commentary and crowd noise. Works with any visual style, photoreal or illustrated, and locks likeness through image references instead of text prompts. Useful if you want to recreate that specific viral aesthetic of someone filming the big screen with their phone. The cost gate is explicit (roughly $15-$35 per generation) and there's a five minute timeout if you spend too long in prep before calling the first paid endpoint.

skills.sh installs: 1.7K
GitHub stars: 41
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add pika-labs/pika-plugins --skill kiss-cam --agent claude-code

Installs into .claude/skills of the current project.

Files

kiss-cam

A two-call pika pipeline: spectator-POV Jumbotron still (gpt-image-2) → 15-second in-arena kiss cam clip with PA-announcer commentary and crowd reaction (kling-v3-omni, first-frame-locked to the still). The trend look is calibrated — pass both reference images straight through. There is no textual substitution into the prompts; the subjects are anchored only through images. Don't reach for ${subject_a} / ${subject_b} placeholders. Step 1 and Step 2 prompts are verbatim, not scaffolds.

Prerequisites

pika MCP available in the host. Tool name prefix varies by mount point — use whatever the host exposes. Tools needed: identity balance, asset upload, image generation, reference-video generation, and async status follow-up when a generation does not complete inline.

Cost transparency gate

Before any paid MCP call, call identity_balance({verbose: true}) once. Surface the current balance, recent burn rate, and remaining runway, then gate the run with this exact message:

SKILL.md on GitHub