baseball-trend

Generates a fake ESPN broadcast still and 15-second Kling video cutaway featuring a user.

1|Updated Jun 20, 2026
One-click install
npx skills add https://github.com/BeanyZoldyck/moggie --skill baseball-trend-beanyzoldyck
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: baseball-trend
Source: https://github.com/BeanyZoldyck/moggie/tree/main/.agents/skills/baseball-trend
Command: npx skills add https://github.com/BeanyZoldyck/moggie --skill baseball-trend-beanyzoldyck

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Creating the viral fake MLB broadcast cutaway trend requires carefully calibrated prompts across two generation engines; this Skill packages the tested recipe so users get a coherent result without prompt engineering. ## Core Features & Use Cases - Broadcast still generation: Uses gpt-image-2 with a reference photo to bake the ESPN scorebug and a name chyron into frame 0. - 15-second video generation: Uses Kling v3-omni with first-frame locking, strict prompt adherence, and native two-announcer commentary that names the user. - Use Case: A user provides their name and a portrait photo, and receives a 16:9 video that looks like a real ESPN cutaway of them sitting behind home plate at a Yankees vs Red Sox ALCS game. ## Quick Start Ask the assistant to make a behind-home-plate broadcast cutaway for a given name using a provided photo URL or file.

Frequently Asked Questions about baseball-trend

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I make a fake ESPN baseball broadcast video of myself?

Provide your name and one front-facing portrait photo. The Skill generates a broadcast still with a scorebug and name chyron via gpt-image-2, then a 15-second Kling v3-omni video with two announcers naming you on air.

What inputs does the baseball broadcast cutaway need?

It needs exactly two inputs: the name as it should appear in the chyron and commentary, and one front-facing or 3/4 portrait photo with good lighting and a single face, supplied as an HTTPS URL or uploadable local file.

Why does the chyron pop in mid-video instead of showing from the start?

The chyron was not baked into the still image at frame 0. Re-run the still generation step and verify the chyron is visible in the image before generating the video, since Kling pixel-locks whatever exists in the first frame.

Can I use a celebrity photo for the fake broadcast trend?

No. Kling's content moderation blocks recognizable celebrity references, returning a risk-control failure at submit time. Use a non-public-figure photo so the chyron name and face remain coherent.

Why use Kling instead of Seedance for this video trend?

Seedance rejects both input images and generated output containing recognizable faces via its partner_validation_failed gate, which every crowd cutaway triggers. Kling is the engine that works for ordinary user photos.

Why does the face change late in the generated video?

Identity drift late in the clip means Kling lost the face, usually because the reference was too small. Re-run the video step, and if drift persists, regenerate the still with a tighter face crop for a stronger identity lock.