Use when the user asks for baseball trend or a task matching the examples below. Viral fake "ESPN behind-home-plate broadcast cutaway" of a user — broadcast-style still + 15s Kling-omni clip with native two-announcer commentary that names the user. Fixed trend: Yankees vs Red Sox ALCS Game 3 at Fenway Park, premium seats, scorebug + chyron with the user's name. Triggers: "make me a behind-home-plate cutaway", "fake MLB broadcast of me", "AI ESPN baseball crowd shot", "viral MLB broadcast trend",
Detected risks:
Baseball Trend is a fixed-recipe Pika Labs plugin skill that produces a viral-style, roughly 15-second 'ESPN behind-home-plate broadcast cutaway' of a user. From just the user's name and one reference photo it generates a broadcast-style still (GPT-image-2) plus a Kling-omni video clip with native two-announcer commentary that names the person on air, set at a fixed scenario: a fake Yankees vs Red Sox ALCS Game 3 at Fenway Park, premium seats, with a scorebug and chyron carrying the user's name.
The skill is a calibrated pipeline: it substitutes the username into pre-written prompts and keeps marked anchors intact. It enforces a cost-transparency gate (estimated 3,000-5,500 credits / ~$30-$55) as the only yes/no confirmation, then runs end-to-end. Notably, it includes an avatar-type moderation probe that classifies the reference image before any paid generation and stops when the subject is a recognizable trademarked or copyrighted character (for example a superhero), routing the user to supply a real-looking photo instead. It uses Kling-omni's native commentary rather than agent-side custom voices, and refers users who want custom voices to the podcast skill.
Target users are individuals wanting a fun, shareable novelty clip of themselves, and social/marketing creators chasing the trend. Because the output is a synthetic 'fake broadcast' that places a real person into fabricated footage naming them, it carries mild synthetic-media/likeness considerations, though it is driven by the user's own photo and includes moderation guardrails against copyrighted-character and celebrity references.
Two things: the name exactly as it should appear in the chyron and announcer dialogue, and one front-facing or 3/4 portrait photo (an HTTPS URL, or a local file it uploads to a public URL).
The cost gate estimates about 3,000-5,500 credits (~$30-$55) for the broadcast still, one or two Kling v3-omni 15s renders, and QA. This is the only confirmation gate; after 'proceed' the pipeline runs end-to-end.
No. An avatar-type probe classifies the image first and stops before generation if it detects a trademarked/copyrighted character, since most video providers will moderate and refuse it. You are asked to supply a real-looking photo instead.
Not in this skill. It uses Kling-omni's native two-male-announcer commentary and does not consume custom voice settings; for custom voices you are routed to the podcast skill with a baseball framing.
Only minimally. It is a fixed trend (Yankees vs Red Sox ALCS Game 3 at Fenway); you substitute the username and reference photo while keeping the calibrated prompt anchors.
Quick Setup:
.claude/skills/Repository
pika-labs/pika-plugins