Sep 12, 2026

Direct Camera-Ready Portrait Prompts with Female Portrait Director

Turn a few portrait inputs into one coherent, camera-ready scene across 20 directed styles — or generate the image directly.

#tutorial#creative

Image models reward directors, not keyword dumpers. Female Portrait Director locks your explicit requirements, picks one visual route, and directs subject, action, clothing, scene, camera, lighting, and finish as a single photographable moment — across 20 implemented styles.

Why This Skill Matters

This is not a generic prompt collection, as the README puts it. Instead of stacking conflicting style keywords, the skill routes each request through one on-demand style file, expands short parameters into a coherent photographed moment, and returns a copy-ready prompt with photography-director intent. It also handles e-commerce clothing model images — preserving clothing-display priority — and direct generation that preserves an authorized person or product reference. The repo has 1,499 stars on the SkillMap leaderboard.

Installation

Install with the open-source skills CLI:

npx skills add https://github.com/liyue-aigc/female-portrait-director/tree/main/skills/female-portrait-director -g

Update an installed copy with:

npx skills@latest update female-portrait-director -g -y

Prefer Git on Codex? Clone into your Codex skills directory (~/.codex/skills/female-portrait-director), restart Codex, and invoke $female-portrait-director.

Real Workflow: Generate a Clean Lifestyle Portrait

  1. Install the skill and start a new agent session.
  2. Paste the README's 60-second example:
Use $female-portrait-director to generate an image directly:
Style: clean lifestyle portrait
Scene: quiet cafe window seat in the afternoon
Outfit: ivory knitted cardigan + light inner top
Mood: gentle, natural, clearly adult
Aspect ratio: 3:4
  1. Remove "generate an image directly" and the same input returns a copy-ready prompt instead of an image. Either way the output follows a fixed format: locked parameters, module analysis, final prompt, negative constraints.

A second workflow produces a styled prompt from a fuller parameter sheet. Feed the skill a Gufeng xianxia request — scene, outfit, mood, facial direction, camera framing, lighting, filter, aspect ratio — and it returns the same four-part output: the locked parameters, a module-by-module analysis, the final directed prompt, and negative constraints.

Tips

  • Run $female-portrait-director with no parameters first: the initial invocation displays the V1.6 tutorial — all 20 styles, basic and advanced templates, style-plus-mood combination guidance, and a full parameter-to-prompt example.
  • The 20-style menu spans clean lifestyle, urban fashion, Gufeng xianxia, retro Hong Kong street, French relaxed, and 15 more; the first-use guide (skill/help.md) documents routes and overlay combination rules.
  • The 62-file distribution package is installer-verified for Codex, Claude Code, Cursor, GitHub Copilot, and Gemini CLI. Only Codex was runtime-executed in this release; the other rows are packaging and specification compatibility claims, per the README's own fine print.
  • Direct image generation depends on your host: Codex supports it when image generation is available, while the other agents depend on the connected image tool.

When Not to Use This

The skill defaults text-only generation to fictional, clearly adult subjects, and its safety boundaries prohibit sexualized minors, explicit nudity, non-consensual images, deceptive identity content, harassment, and defamation — read docs/prompt_safety.md before the curve-focused and reference-image workflows. And if you need free-form art direction rather than one locked, directed route, a lighter prompt workflow will fight this skill's constraints.


See the leaderboard for more skills.