cinematic image to video director
Directs cinematic AI video generations with precise technical specs, camera movements, and continuity checks.
$5
Works with the AI Agents you already use
Claude Code
Cursor
Codex CLI
Muse
Dots+21 moreClaude CodeCursorCodex CLIMuseDots
Manus
Grok
Grok Bot+18 agentscinematic image to video director
Example session with this skill installed
Direct a shot based on this idea: A futuristic detective walks through a neon-lit rain-slicked alley, stops, and looks up at a glowing hologram billboard. It's the second shot in the scene.
- Read your context and instructions
- Compiled the cinematic image to
I have developed the production specifications for Shot 2, ensuring continuity with the detective's established wardrobe and the alley's low-key lighting.
Production Specification: The Neon Alley
STORY INTENT: Discovery — the detective finds the target of their search in a moment of overwhelmed realization.
SHOT: Medium Long Shot (MLS) |
LENS: 35mm (Natural) |
MOVEMENT: Handheld follow, transitioning to a slow tilt-up.
BLOCKING:
- Start: Detective enters from bottom-right, walking away from camera.
- End: Detective stops center-frame, looking up at a 60-degree angle.
- Path: Linear movement along the Z-axis (away), then stationary.
LIGHTING: High-contrast cyan/magenta rim lighting, soft top-down ambient |
ATMOSPHERE: Heavy rain, volumetric steam from floor vents, lens flares from neon signs.
TIMING:
- 0:00-0:03: Handheld follow of the detective walking; rain splashes on pavement.
- 0:03-0:05: Detective slows to a halt; camera settles into a static position.
- 0:05-0:08: Detective tilts head upward 60°; camera tilts up simultaneously to reveal the billboard.
DIALOGUE/SOUND: Heavy rhythmic rain, distant synth hum, wet boot splashes on asphalt.
CONTINUITY: Matches Shot 1; Detective's black trench coat and wet hair texture are maintained. Lighting direction remains consistent from camera-left.
PROMPT:
Cinematic MLS, 35mm lens. A futuristic detective in a black wet trench coat walks away through a dark neon-lit alley. Cyan and magenta rim lighting, heavy rain, volumetric steam. At 3 seconds, they stop walking center-frame. At 5 seconds, they look up toward a glowing hologram billboard. Handheld follow camera transitions to a slow tilt-up. High detail, 4k, consistent lighting, realistic fluid motion.
Next steps
- Copy the PROMPT into Runway Gen-3 or Sora to generate the clip.
- Review the tilt-up transition for any subject morphing.
- Provide the end-frame of this shot if you need to generate Shot 3.
Connects securely to your tools. The creator never sees your data.
About this skill
The problem
Static image-to-video tools often produce melting subjects, unpredictable speed ramps, or "drift" where lighting and characters change mid-shot. Developers and creators waste expensive credits on generations that ignore camera physics or break visual continuity in a sequence.
What it does
- Analyzes reference images or shot descriptions to extract subjects, lighting, and composition.
- Builds precise technical specs including lens focal lengths (14mm to 135mm) and specific camera movements like orbits or dollies.
- Generates timecoded motion breakdowns to prevent the AI from "guessing" event duration.
- Provides a continuity engine that checks new shots against previous end-frames to maintain wardrobe and lighting consistency.
- Diagnoses failed generations by mapping visual artifacts to specific prompt fixes.
Frameworks & tools
Optimized for high-end video models including Sora, Runway Gen-3 Alpha, Pika, Kling, and Veo.
Why this beats prompting it yourself
Instead of hoping the AI understands "cinematic," this skill enforces professional cinematography standards. It ensures camera movement is physically plausible and provides the structured timing data required to keep video models from hallucinating motion paths.
Use cases
- Converting concept art into a consistent 5-second cinematic reveal.
- Generating a multi-shot sequence where characters and settings must match across clips.
- Fixing "melting" or morphing artifacts in failed AI video generations.
- Scripting complex camera work like slow-dolly-ins with 45-degree head turns.
Known limitations
Does not render video files or handle live-action scheduling, budgeting, or casting. Requires a separate video generation tool to execute the final prompt.
No video rendering capability. Cannot manage production budgets or casting. Focuses exclusively on technical direction and prompt construction.
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 any payment.
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 1 month ago
Permissions
No special permissions declared or detected