Skip to main content
skill CX Codex

gameday — Claude Code skill: 6 matchday fan photos

Turns one photo of you into six stadium shots in your team's kit. Needs the Pika MCP.

Install into Claude Code · Skills CLI (Vercel) All methods and platforms ↓
npx skills add Pika-Labs/Pika-Plugins --skill gameday -a claude-code
Open on GitHub

What gameday does

The skill takes a single photo and a team name, then returns six separate vertical images of you at the game in that team's jersey and scarf. The set follows a matchday arc: arriving at the stadium gates, sitting in the seats before kickoff, laughing with a friend in the stands, the goal celebration with both arms up, a crowd selfie, and a walk along the concourse with a flag over your shoulders.

Six prompts ship with the skill. Each one goes out as its own generate_image_edit call with the same source photo passed as images, which is what keeps the face consistent across the carousel. Every call uses provider: gpt-image-2, aspect_ratio: 3:4, quality: medium and output_format: png. A local file is uploaded first through upload_asset and a PUT to presigned_url, then the public_url goes into the prompts.

Kit colours are described in words, not crests or logos. If your club has a busy pattern, spell it out yourself: two main colours plus the stripe type. The generation calls fail without the Pika MCP, so wire that up first.

What it helps with

  • produce six matchday shots from one uploaded photo
  • keep the same face across every image in the set
  • build a social carousel in your team's kit
  • get vertical 3:4 photos without cropping by hand
  • describe club colours in text instead of using logos

Example prompts for the agent

  • /gameday ~/photos/me.jpg Argentina
  • make me an Argentina superfan from this photo
  • put me at the game, here's my picture and team

How to install gameday

1Skills CLI (Vercel)

npx skills add Pika-Labs/Pika-Plugins --skill gameday -a claude-code

2Manually: put the files in place

~/.claude/skills/gameday/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/gameday/

Step by step, with commands for each platform: How to install skills in Claude Code: five ways

Questions about gameday

How do I install gameday in Claude?

Run npx skills add Pika-Labs/Pika-Plugins --skill gameday -a claude-code. The skill lands in ~/.claude/skills/gameday/SKILL.md.

What does gameday do?

From one photo and a team name it generates six separate images of you as a fan at the stadium in that team's kit. Together they form a matchday carousel.

What do I need for the skill to work?

The Pika MCP connected, plus a photo as a local path or URL. Local files are uploaded by the skill before generation.

Source text

SKILL.md · Apache-2.0
---
name: gameday
description: >-
  Use when the user asks for gameday or a task matching the examples below.
  Put a fan into the game. The user uploads one photo of themselves and names
  their favorite team; generate a 6-photo matchday-superfan carousel — six
  separate images of that same person at the game in their team's kit. Triggers:
  "/gameday", "make me an [team] superfan", "put me at the game", "gameday
  carousel". Requires the Pika MCP.
argument-hint: <user-photo-path-or-url> <favorite-team>
---

# gameday

The user uploads **one photo of themselves** and gives their **favorite team**. You generate **6 separate photos** of that person as a superfan at the game — a matchday carousel.

## Steps

1. **Get the two inputs.** Photo (local path or URL) and favorite team. If the photo is a local path, upload it (`upload_asset` → PUT to `presigned_url` → use `public_url`). Call the resulting URL `photo`.

2. **Fill the team.** From the team name, fill `{team}` and `{kit colors}` (two main colors + stripe pattern, e.g. Argentina → "sky-blue and white vertical stripes"). Describe colors, not crests/logos.

3. **Generate the 6 photos.** Make **6 separate `generate_image_edit` calls**, one per prompt below, passing the same `photo` as `images` on every call (this keeps the face identical). Settings each call: `provider: gpt-image-2`, `images: [photo]`, `aspect_ratio: 3:4`, `quality: medium`, `output_format: png`.

4. **Deliver the 6 separate images** in order. Six files — never one image split into panels.

## The 6 prompts (substitute `{team}` / `{kit colors}`)

**p1 — Arrival at the stadium**
```
Ultra-realistic candid lifestyle sports-fan photograph, single vertical phone photo, 3:4. The
subject is the SAME person as the reference photo — preserve their exact face, hair, features,
skin tone and vibe; natural candid expression. A passionate {team} football supporter arriving on
match day, wearing an official {team} home jersey ({kit colors}) with a matching {kit colors}
{team} supporter scarf looped around the neck, effortless game-day styling. They walk confidently
toward a big modern stadium's entrance gates, golden late-afternoon sun flaring low behind the
stands, long warm shadows. Streams of other fans in the same {kit colors} flow around them,
flags and team colors everywhere, a buzzing pre-match crowd. Candid street-style framing, slight
motion in the step, shallow depth of field on the background, realistic smartphone photo look,
authentic and unposed.
```

**p2 — In the seats before kickoff**
```
Ultra-realistic candid lifestyle sports-fan photograph, single vertical phone photo, 3:4. The
subject is the SAME person as the reference photo — preserve their exact face, hair, features,
skin tone and vibe. A {team} supporter in an official {team} jersey ({kit colors}) and matching
{team} scarf, sitting in the stadium seats before kickoff, leaning casually on the armrest, looking
thoughtfully out toward the green pitch. Behind them a massive blurred crowd fills the stands,
a sea of {kit colors}, the floodlights and tiered stadium bowl rising up. Soft natural daylight,
candid quiet-before-the-storm mood, gentle bokeh on the crowd, realistic smartphone photo look,
authentic and unposed.
```

**p3 — Laughing with a friend in the stands**
```
Ultra-realistic candid lifestyle sports-fan photograph, single vertical phone photo, 3:4. The
subject is the SAME person as the reference photo — preserve their exact face, hair, features,
skin tone and vibe. A {team} supporter in an official {team} jersey ({kit colors}) and matching
{team} scarf, standing in the stands smiling and laughing mid-conversation with a friend beside
them, a joyful candid fan moment caught off-guard. A packed, colorful crowd in {kit colors} rises
behind them, banners and flags waving, lively pre-match energy. Bright stadium daylight, genuine
happy expression, slight candid blur, realistic smartphone photo look, authentic and unposed.
```

**p4 — Goal celebration (hero shot)**
```
Ultra-realistic candid lifestyle sports-fan photograph, single vertical phone photo, 3:4. The
subject is the SAME person as the reference photo — preserve their exact face, hair, features,
skin tone and vibe. A {team} supporter in an official {team} jersey ({kit colors}) caught at the
exact moment of a goal — standing in the stands cheering, both arms thrown high, the {kit colors}
{team} scarf stretched wide overhead with both hands, mouth open in an ecstatic roar of pure
euphoria. All around them the crowd erupts, fans jumping and screaming in {kit colors}, streamers
and confetti in the floodlit air, electric chaotic energy. Dynamic candid framing, slight motion
blur from the celebration, high contrast stadium lights, realistic smartphone photo look,
authentic and unposed.
```

**p5 — Crowd selfie close-up**
```
Ultra-realistic candid lifestyle sports-fan photograph, single vertical phone photo, 3:4. The
subject is the SAME person as the reference photo — preserve their exact face, hair, features,
skin tone and vibe. A close-up selfie-style portrait of a {team} supporter in the packed crowd,
wearing an official {team} jersey ({kit colors}) with the {kit colors} {team} scarf draped behind
their shoulders, giving a big genuine smile straight to camera, cheek-to-cheek with the roaring
stands behind. Bright stadium floodlight bokeh and a blurred sea of fans in {kit colors} fill the
background, evening match atmosphere. Slight selfie wide-angle look, warm skin tones, realistic
smartphone photo, authentic and unposed.
```

**p6 — Flag-draped on the concourse**
```
Ultra-realistic candid lifestyle sports-fan photograph, single vertical phone photo, 3:4. The
subject is the SAME person as the reference photo — preserve their exact face, hair, features,
skin tone and vibe. A {team} supporter walking along the open stadium concourse with a large
{kit colors} {team} flag draped over their shoulders like a cape, wearing the official {team}
jersey ({kit colors}) underneath, proud triumphant expression, chin up. Warm golden-hour light
pours through the concourse openings, lens flare, fellow fans in {kit colors} milling all around,
the stadium structure framing them. Candid editorial street-style framing, cinematic warm tones,
realistic smartphone photo look, authentic and unposed.
```

© Pika-Labs · Apache-2.0 · Pika-Labs/Pika-Plugins

Similar skills

Everything for AI and ML →

From the same set12

Project sponsors

Companies that keep this analytics open
Gold sponsor

Impressions, clicks and code copies — open statistics. We keep the numbers public for the same reason NeBlask itself is public. Become a sponsor