AGENT SKILL

varg-character-beautiful-woman

Makes very beautiful AI women who look like real models, not plastic dolls, on varg: striking features written one by one, real skin, styling (hair, makeup, outfit) and light that make a photo look like a fashion, swimwear or red-carpet shoot, plus the wording that gets these prompts past false moderation blocks. Adults only, no nudity. Use when someone asks for a beautiful, gorgeous, attractive or hot woman, a model, an AI influencer girl, a glamour or swimwear shot, or a woman who looks real and stunning. Part of varg-character-design (definition, sheet, reuse in video).

License 许可
MIT
最近更新
2026-11-11
Stars
★ 1.6k
Forks
474

01 前言 Front Matter

name
varg-character-beautiful-woman
license
MIT
metadata
{ "author": "vargHQ", "version": "1.0.0", "openclaw": { "homepage": "https://varg.ai" } }
description
Makes very beautiful AI women who look like real models, not plastic dolls, on varg: striking features written one by one, real skin, styling (hair, makeup, outfit) and light that make a photo look like a fashion, swimwear or red-carpet shoot, plus the wording that gets these prompts past false moderation blocks. Adults only, no nudity. Use when someone asks for a beautiful, gorgeous, attractive or hot woman, a model, an AI influencer girl, a glamour or swimwear shot, or a woman who looks real and stunning. Part of varg-character-design (definition, sheet, reuse in video).
allowed-tools
Bash(curl:*) Read Write Edit
compatibility
Needs the varg MCP connection (https://mcp.varg.ai/mcp) or a VARG_API_KEY (see varg-ai).

03 SKILL.md 正文

Very beautiful women

Two ways this goes wrong. "Beautiful woman, 8k" gets the model's average face: pleasant and forgettable. "Hot sexy girl" gets blocked on strict models, and where it passes it gives the glossy, poreless, same-face "AI girl" that nobody believes. What works is a model-casting description plus fashion styling plus real skin, framed as a shoot.

Sheet, crops and video: varg-character-design. Scope: adult women (say the age), clothed (fashion, evening wear, swimwear, sportswear), no nudity, no sexualised posing, no real person's likeness.

The recipe

1. Cast her like a model agency would. Age as a number (22–35) and 4–6 striking features:

  • Bone structure: "high cheekbones", "a sharp jawline", "a strong straight nose"
  • Eyes: "almond-shaped green eyes", "striking ice-blue eyes", "large dark eyes with long lashes", "thick arched brows"
  • Lips: "full lips", "full glossy lips"
  • Hair, with a styling verb: "long dark glossy hair in soft waves", "sleek platinum-blonde hair", "sun-bleached hair slicked back and wet", "voluminous dark curls"
  • Skin tone and origin, specifically: "sun-kissed olive skin", "bronzed skin", "fair skin with a light flush"

2. Keep the skin real. "Visible pores, a few freckles, fine peach fuzz, real skin texture, no plastic smoothing." Beauty with real skin is what makes people stop and believe.

3. Style her. Makeup ("glowing natural makeup with defined lashes and a nude-pink lip", "a bold red lip", "cat-eye eyeliner, glossy lips"), one piece of jewellery (gold hoops, a thin gold necklace), and an outfit named exactly:

  • Evening: a fitted black satin slip dress; an emerald satin gown with a high leg slit
  • Swim: a cherry-red one-piece swimsuit with a high leg cut (name the cut; never "skimpy")
  • Night out: a fitted black mini dress (not "tight")
  • Day: a white linen shirt open at the collar

4. Light and lens make it a shoot. Golden hour with warm backlight in the hair, 85mm, shallow depth of field; or a direct-flash compact-camera look for a candid party photo; or editorial flash on a red carpet.

5. One confident action: leaning on a balcony railing, glancing back over her shoulder, laughing with a drink, posing at a premiere.

Ready characters (all tested on 2026-10-27)

Look Model Prompt (shortened) Result
Positano editorial seedream_5_pro (7) strikingly beautiful 26-year-old woman with model looks: high cheekbones, full lips, almond-shaped green eyes, sharp jawline, long dark glossy hair in soft waves, sun-kissed olive skin with visible pores and a few freckles, glowing natural makeup; fitted black satin slip dress, thin gold necklace, leaning on a balcony above the sea at golden hour, confident half-smile, 85mm, warm backlight, no plastic smoothing image
Red carpet gpt_image_2_5 (24) strikingly beautiful 29-year-old woman, sleek platinum-blonde hair, sharp cheekbones, ice-blue eyes, bold red lip, real skin texture; fitted emerald satin evening gown with a high leg slit, long gold earrings, film premiere, photographers' flashes, full length, editorial flash image
Beauty close-up nano_banana_2_1 (6) gorgeous 25-year-old Brazilian woman, high cheekbones, full glossy lips, large dark eyes, thick arched brows, bronzed skin with visible pores, voluminous dark curls, gold hoops, white linen shirt, golden-hour sun, 85mm f/2, magazine cover quality image
Party, direct flash soul (6) very attractive 24-year-old woman at a rooftop party at night, long straight black hair, cat-eye eyeliner, glossy lips, small nose ring, fair skin with visible pores; fitted black mini dress, silver hoops, laughing with a drink, city lights, compact-camera flash, slight grain image
Swimwear campaign seedream_5_pro (7) swimwear brand campaign photo, stunningly beautiful 27-year-old adult woman, athletic figure, wet slicked-back sun-bleached hair, golden tanned skin with pores and water droplets, blue eyes; cherry-red one-piece swimsuit with a high leg cut, knee-deep in turquoise water at sunset, glancing back with a confident smile, Kodak Portra look image

Best for striking glamour with real skin: seedream_5_pro. Best for a believable candid: soul or nano_banana_2_1.

Getting past false blocks

Instead of Write Why (tested)
girl, young girl woman, 26 / adult woman, 26 "Beautiful girl in a sports bra" was blocked on GPT Image 2.5
sexy, hot, seductive strikingly beautiful, confident, model looks crude words get blocked or give the plastic look
tight dress fitted black mini dress same picture, no trigger
bikini the cut + "swimwear brand campaign" "bikini" was blocked on GPT Image 2.5, FLUX 3 and Grok
booty, cleavage the pose and the neckline: "glancing back", "square neckline" describe the photo, not body parts

Route swimwear to seedream_5_pro, soul or nano_banana_2_1; GPT Image 2.5 and Grok Image 2 block most swimwear. Retry a block once (they are random), then rephrase. Full log: varg-character-design → references/moderation.md.

Keep her the same

Make her model sheet with seedream_5_pro from the same casting description (the sheet prompt is in varg-character-design), then every post from the face and body crops. For an AI influencer, fix one signature: a nose ring, a beauty mark, one hair colour, one jewellery piece.

Watch out

  • The more generic the description, the more the face drifts toward a celebrity; some filters then block it as likeness. Specific features prevent both.
  • Label AI models and influencers as AI on Instagram and TikTok; never present her as a real customer in an ad.

04 同仓库更多 Skills

hashgraph-online
hashgraph-online/awesome-codex-plugins

migrate-to-e2e

Moves an existing Playwright, Cypress, Selenium, Appium, Detox, Maestro, or TesterArmy suite to e2e one slice at a time, with the old suite running beside the new one until each ported test is proven. Keeps a MIGRATION.md progress file so any later session resumes where the last stopped. Optionally plants a defect to show both suites catch it. Use for /migrate-to-e2e, "migrate our tests to e2e", "port these Maestro flows", "move our TesterArmy tests", or when a project has an e2e config and an older suite still running.

★ 1.6k
hashgraph-online
hashgraph-online/awesome-codex-plugins

e2e

Writes, runs, and debugs end-to-end tests for web and mobile apps with e2e, the agentic test runner, which mixes agent.act goals with exact locator assertions on a Playwright browser or an iOS or Android device. Also drives a project's running app live over the e2e MCP server to look at a screen, check a UI change, take a screenshot or record a video, and try a locator; and runs bug bashes, parallel e2e explore runs with each bug proven by a repro test. Use when a project depends on e2e or has an e2e.config.ts, when asked for end-to-end, browser, mobile, or agentic UI tests, to check a change in the running app, to QA or bug bash an app or branch, or when an e2e run fails.

★ 1.6k
hashgraph-online
hashgraph-online/awesome-codex-plugins

create-verification-skill

Generate a project-local verify-<app> skill on top of e2e, so any coding agent can launch the app, check it is worth driving, exercise a feature the way a user does, keep evidence, and bug bash it. Writes the skill, a feature map with one explore charter per feature, and proves it once. Use for /create-verification-skill, "make a verification skill for this repo", or when a project has no scripted way to prove UI behavior.

★ 1.6k
hashgraph-online
hashgraph-online/awesome-codex-plugins

deja-search

deja-vu memory — search the user's past AI coding sessions with the deja CLI. Holds what they already settled and never wrote into the repo (naming rules, banned or already-tried tools, required checks before merge, preferred values, who owns what). Use before your first edit in a task, before you call a change done or ready to merge, before you recommend a tool, library, design or reviewer, before re-debugging an error that may already be solved, when they say 'didn't we fix this before', 'what did we decide about X' or 'I already have that', and before telling them something on this machine does not exist.

★ 1.6k
hashgraph-online
hashgraph-online/awesome-codex-plugins

deja-reflect

End-of-session reflection and memory curation with deja-vu. Review what was decided, what commands broke and what repaired them, and record durable project rules or decisions with deja remember so the next session starts with them.

★ 1.6k
hashgraph-online
hashgraph-online/awesome-codex-plugins

revid

Make, edit, export and post AI videos with Revid (revid.ai) through the Revid MCP server. Use whenever the user wants a short-form or long-form video made (from an idea, a script, a URL, an article, a PDF, a tweet, a product page, their own clips, or a song), wants to find what is going viral on TikTok, YouTube, Instagram or X before making something, wants to edit or re-export an existing Revid project, generate a standalone AI image or clip, make a series episode, or publish and schedule a finished video to connected social accounts. Covers setup (hosted MCP with OAuth, or an API key), the 36 tools, credits, polling and the rules for publishing. Routes to the more specific revid-* skills when one fits.

图像与多媒体Skill与智能体开发
★ 1.6k