- Home
- Skills
- Image & Video
- Cinematic Shot Prompting
Works with the AI Agents you already use
Cinematic Shot Prompting
Convert vague narrative ideas into precise 7-slot camera motion prompts for AI video generators.
$3.99
Cinematic Shot Prompting
Example session with this skill installed
I have a reference image of a woman sitting in a cafe. I want a 5-second shot that feels lonely and focuses on her expression as we get closer. Help me write the prompt for Kling.
- Read your context and instructions
- Compiled the cinematic shot prompting
The narrative goal is to enhance emotional intimacy by transitioning from a medium shot to a tight close-up on the subject's eyes.
Cinematic Motion Contract
Motion Selection: Slow Dolly-In (Forward Translation)
Prompt:
[Camera Instruction]: Start with a medium shot, slow and steady dolly-in toward the subject's face, linear velocity, ending in a tight close-up on the eyes. [Subject]: The woman remains seated and mostly still, maintaining a melancholy gaze, slight blinking permitted, facial features must remain anatomically stable. [Environment]: Background cafe bokeh remains consistent, visible parallax shift in the foreground table edge as the camera advances. [Constraints]: Maintain focus on the pupils, no sudden camera shakes, background lighting remains constant.
Keyframe Suggestions:
- Frame 0s (Start): Use your provided reference image (Medium Shot).
Frame 5s (End): Generate or prompt a close-up image of the same woman’s face with identical lighting and features to act as the terminal anchor.
Next steps
- Test this prompt with a 5-second duration in Kling.
- If the face deforms during the zoom, provide the 5s terminal keyframe to lock the geometry.
- If the background "breathes," add a constraint for "fixed focal length" to distinguish the dolly move from a digital zoom.
Connects securely to your tools. The creator never sees your data.
About this skill
Video generation models like Kling, Runway, and Luma often produce erratic camera movements, deformed subjects, or background jumps when given vague "cinematic" prompts. This skill stops the drift by compiling narrative intent into precise, physically-grounded camera motion contracts.
What it does
- Compiles motion contracts by defining start/end compositions, geometric paths, and velocity curves.
- Isolates camera mechanisms to prevent conflicting instructions like simultaneous fixed-position and orbiting moves.
- Enforces anti-drift constraints that keep subjects stable in the frame while the background moves.
- Plans keyframe strategies for high-risk transitions such as window fly-throughs or macro-to-micro zooms.
- Diagnoses motion failure by mapping glitches like "jittery backgrounds" to specific slot corrections in the prompt grammar.
How it works
- Identify the input mode based on whether you have a reference image or just a text description.
- Select the primary mechanism such as pedestal, pan, tilt, zoom, or tracking.
- Draft the motion contract covering camera instructions, subject stability, and environmental parallax.
- Determine keyframe requirements for complex spatial transitions that a single prompt cannot bridge.
Frameworks & tools
Optimized for Kling, Runway Gen-3/Gen-2, Sora, and Luma Dream Machine. Works with image-to-video and text-to-video workflows.
Why this beats prompting it yourself
Generic prompts rely on the model's "artistic" interpretation, which leads to unpredictable results. This skill uses a 7-slot structural approach that treats camera movement as a verifiable physical instruction, significantly reducing compute waste on failed generations.
Use cases
- Emotional close-ups by calculating a slow push-in that keeps facial features stable.
- Spatial establishing shots using parallax-heavy lateral tracking to reveal depth.
- Complex transitions like flying a camera through a narrow opening into a new scene.
- Repairing jittery output by isolating and fixing conflicting motion vectors in an existing prompt.
Known limitations
This skill focuses on camera and focal movement. It does not generate aesthetic style descriptions, character designs, or lighting prompts from scratch.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Video Skills.
Focuses strictly on camera/focal motion. Does not design character aesthetics, lighting, or static art styles. Best results require existing reference images for visual consistency.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 13 days ago
Permissions
No special permissions declared or detected