Realistic Profile Pictures — AI skill for Claude Code
Agent skill (Claude Code, Cursor) that generates realistic, non-AI-looking profile pictures and avatars for mockups, websites, marketing and video.
How to install Realistic Profile Pictures
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open heyimjames/realistic-profile-pictures and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Realistic Profile Pictures does
Agent skill (Claude Code, Cursor) that generates realistic, non-AI-looking profile pictures and avatars for mockups, websites, marketing and video. RealPFP prompt engine + fal.ai nano-banana-2, bring your own key.
Alternatives in AI
- Skill Seekers Roadmap — Transform Skill Seekers into the easiest way to create Claude AI skills from any knowledge source - documentat 11.1k ★
- Bot On Anything — A large model-based chatbot builder that can quickly integrate AI models (including ChatGPT, Claude, Gemini) i 4.2k ★
- Cumora — Where agent teams gather 3.2k ★
README
Realistic Profile Pictures
Generate realistic AI profile pictures and avatars that **do not look AI-generated**. For filling mockups, websites, marketing, demos, and video with believable, diverse people, instead of uncanny AI output or the same recycled stock faces.

Every face below is generated. None of these people exist.

Packaged from [realpfp.vercel.app](https://realpfp.vercel.app) · source: [heyimjames/RealPFP](https://github.com/heyimjames/RealPFP)
Install
Works with Claude Code, Cursor, and any agent that supports [skills](https://www.npmjs.com/package/skills):
npx skills add heyimjames/realistic-profile-pictures
Then just ask for "realistic profile pictures", "avatars for a mockup", or "placeholder faces".
Run it directly
Needs Node 18+ and a free [fal.ai key](https://fal.ai/dashboard/keys) (bring your own).
# preview prompts only, no API calls, no cost
node scripts/generate.mjs --count 3 --mode profile --dry-run
# generate images
FAL_KEY=your_key node scripts/generate.mjs --count 6 --mode profile --out ./faces
| Flag | Values |
|---|---|
--mode |
aspirational (polished), profile (natural), candid (documentary) |
--count |
1 to 50 |
--resolution |
1K, 2K |
--out |
output folder (default ./pfp-out) |
--dry-run |
print prompts, generate nothing |
Why the faces look real
The value is the prompt engine, not the API call. nano-banana-2 is a Gemini-family model, so it:
- Describes what a real photo has (pores, one light source, one catchlight, correct anatomy) instead of naming what to avoid. Saying "no plastic skin" can summon plastic skin.
- Avoids beauty-filter trigger words (flawless, perfect, radiant) that plasticise the result.
- Varies everything that AI tends to fix: body shape, teeth, eyes, hair, wardrobe, background.
More detail: [reference/p
Related Skills
Nano Banana 2 Skill
AI image generation CLI powered by Gemini 3 Pro. Green screen transparency, reference images, style transfer.
Visual Skills
AI film director skills for agents: cinematic dramaturgy (Murch, blocking, montage) + exact prompt syntax for
Nano Banana
Generate or edit images using Google's Gemini Image model
Claude Drives Gemini
Let your AI agent drive the Gemini web app through your existing Google AI Pro subscription - generate images
Img
Gera imagem por IA (gpt-image-2 ou nano-banana-2) via /image-gen
Flow Py
The unofficial Python API, CLI, MCP Server & Universal Skill for Google Flow. Deterministic image (Nano Banana
Related Agents
Asset Generator
AI asset generation agent — connects to image generation APIs (Nano Banana, FAL AI, Replicate), manages asset
Scene Prompt Smith
Use this agent when drafting kie.ai Nano Banana 2 image prompts for the scenes of a short video. It turns a sc
Image Craft Expert
Crafts optimized text-to-image prompts and generates images using Gemini (nano-banana) and/or ChatGPT (gpt-ima