More screenshots
Works with the AI tools you already use
Claude Code
Cursor
Codex CLI
Muse
OpenClaw+21 moreClaude CodeCursorCodex CLIMuseOpenClaw
Manus
Grok
Grok Bot+18 agentscinematic director
Turn scene descriptions into professional cinematic photography prompts using director-level workflows and DP logic.
$9.99
cinematic director
Example session with this skill installed
A 23-year-old female swordsman standing in a bamboo forest at night, holding a sword, looking at the blade. Mid-shot, cinematic wuxia style. No sky visible.
- Read your context and instructions
- Compiled the cinematic director
The scene analysis is complete. We are capturing a moment of quiet reflection, prioritizing the silence of the forest over the action of the martial artist.
===================
Creative Meeting
Director 这一幕真正拍的是：人与剑之间的沉默（主题：回忆与宿命）
Story 她刚刚停下脚步，在月光下确认剑刃的冷冽
Cinematographer 85mm 焦段 / 人物占比 35% / 机位与眼平齐 / 严禁拍摄天空
Art Director 色板：翡翠绿、月光银、炭黑、柔白
Visual Hook 第一眼看剑刃上的月光反光，第二眼看她低垂的眉眼
Attention 她的注意力完全在剑身上，对摄影机毫无察觉
Mood 清冷、静止、悬置
===================
Modules
Character DNA: A 23-year-old female wanderer with a calm, distant temperament. Her features are marked by natural asymmetry, creating a memorable individuality rather than a generic doll-like face. She wears simple, heavy-textured linen robes.
Attention: Her attention is entirely absorbed by the horizontal blade. She never notices the camera, creating a sense of a private, witnessed moment.
Story Moment: The world is suspended. The bamboo stalks are motionless. She has just stopped her movement to observe the steel.
Photography Strategy: 85mm lens at eye level. The camera is tucked within the foliage, making the composition feel discovered. Light source is filtered moonlight through a thick canopy; no direct moon disc is visible.
Visual Hook: A narrow ribbon of silver light caught on the sword's edge acts as the primary visual anchor.
Photographer's Notes: Observe the silence before observing the character. Trust the natural fall of light. Leave large areas of darkness untouched.
Director's Observation: The entire forest seems to hold its breath. The landscape remains indifferent to her presence.
=========================
FINAL PROMPT
This image is approached as a quiet cinematic observation rather than a fantasy illustration. The portrait captures a fleeting moment of contemplation. A 23-year-old wandering Chinese swordswoman stands among dense bamboo at night. She occupies roughly one-third of the frame. Her right hand holds a straight sword horizontally; she quietly watches the cold moonlight reflecting off the steel. Her attention belongs entirely to the blade. She never notices the camera. Soft silver light filters naturally through layers of bamboo leaves, wrapping the scene without revealing its source. The moon itself remains hidden above the canopy. 85mm lens, eye-level perspective. Colors are limited to emerald, silver, and charcoal. Restrained highlights, natural atmospheric perspective, and low contrast. Observe the silence. Trust the landscape.
=========================
NEGATIVE PROMPT
[AI & CG] HDR, oversharpened, CGI, game render, digital painting, plastic skin, Unreal Engine, illustration, anime.
[Composition] centered composition, hero pose, poster layout, perfect symmetry, looking at camera.
[Lighting & Color] neon colors, oversaturated, theatrical lighting, studio flash, overexposed, fake rim light.
[Scene Specific] visible sky, visible moon, stars, clouds, glowing sword, magic effects, energy particles.
[Artifacts] watermark, text, logo, extra limbs, deformed hands.
Connects securely to your tools. The creator never sees your data.
What you get
About this skill
Mainstream AI image generators often produce "game-like" CG renders with plastic skin, flat lighting, and cluttered compositions. This skill stops the agent from acting like a prompt engineer and starts forcing it to act like a cinematic director. It shifts the focus from listing objects to defining photographic intent, atmospheric depth, and the "quiet moments" that define real cinema.
Instead of generic keywords, it produces structured photographer's notes that govern lighting, lens selection, and attention hierarchy. It employs a multi-role workflow—Director, Cinematographer, Art Director—to ensure every output has a distinct visual hook and a grounded, physical sense of space.
What it does
- Multi-Role Simulation conducts a creative meeting between director, DP, and art director to define the scene's emotional core.
- Character DNA Logic designs recognizable, non-perfect characters with specific temperament-feature contrasts rather than generic beauty.
- Attention Mapping anchors the character's focus on internal thoughts or specific objects to create narrative tension.
- Photography Strategy dictates lens focal length (e.g., 85mm eye-level), spatial distribution, and light source logic.
- Negative Constraint Engine filters out "AI-isms" like HDR, over-sharpening, and plastic textures to preserve filmic texture.
How it works
- Analyze the Theme to identify the underlying emotion (loneliness, awe, anticipation) rather than just the objects.
- Assign Photography Type like Environmental Portrait or Landscape and load the corresponding technical template.
- Conduct Creative Meeting to define specific lighting, color palettes, and the "Visual Hook."
- Synthesize Final Prompt into a cohesive English prose narrative following a strict hierarchical structure.
Frameworks & tools
Optimized for Midjourney (v6+), Flux.1, SDXL, and proprietary models like Doubao/Jimeng. It uses a modular knowledge base for wuxia, xianxia, portraiture, and architectural cinematography.
Why this beats prompting it yourself
Most users add "cinematic" or "8K" which actually triggers low-quality training data. This skill uses industry-standard film terminology and negative space constraints that most users overlook, resulting in images that look like stills from a high-budget production rather than a stock photo.
Use cases
- Generating atmospheric concept art for film pre-production.
- Creating high-end fashion or character portraits with naturalistic lighting.
- Producing "Wuxia" or "Xianxia" scenes that avoid mobile-game aesthetics.
- Designing still-life photography for mood boards with specific material textures.
Known limitations
Does not support technical asset generation like 3-view orthographic sheets or product packshots. Requires the agent to have file-reading capabilities enabled for full template access.
How to install
Works the same in every agent - Claude, Cursor, Codex, Copilot and 20+ more.
- 1
Download the ZIP
Free skills download straight away. Paid skills unlock right after purchase.
- 2
Unzip into your skills folder
Every agent reads skills from one folder on your machine. Drop the unzipped folder in there.
- 3
Ask your agent to use it
Restart the agent if it was already running. It picks the skill up automatically - no config needed.
Skills folder by agent
Click the path to copy it. Create the folder if it does not exist yet.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill passes our trust checks below.
Security scanned
Passed our 8-point scan before listing
Fresh listing
Recently published to Agensi
30-day refund
Not a fit? Get your money back
Trust & safety
Security scanned
Verified clean 1 day ago
- Passed all security checks, Safe to install