Works with the AI Agents you already use
Claude Code
Cursor
Codex CLI
Muse
Dots+21 moreClaude CodeCursorCodex CLIMuseDots
Manus
Grok
Grok Bot+18 agentsBlender Previs
Generates low-poly Blender previs and motion scripts to validate blocking, camera paths, and shot timing.
$4
Blender Previs
Example session with this skill installed
Create a previs for a 5-second shot: Character A walks from a door to a desk while the camera follows in a low-angle tracking shot. Audio is 4.8s. Use 30fps.
- Read your context and instructions
- Compiled the blender previs
I have analyzed the shot requirements and generated a previs plan to validate the movement and camera path. The setup uses a 150-frame timeline to accommodate the 5s duration at 30fps.
Previs Construction Plan
| Item | Specification |
|---|---|
| Actors | Character A (Blue Capsule), Desk (Gray Box) |
| Space | 4m x 5m room, door at (0,0,0), desk at (3,4,0) |
| Motion | Constant walking speed from door to desk, stopping at frame 144 |
| Camera | 24mm lens, 0.5m height, tracking parallel to A's path |
| Output | previs.blend, buildPrevis.py, previs.mp4 |
import bpy
def setup_scene():
# Clear existing mesh data in the scene
bpy.ops.object.select_all(action='SELECT')
bpy.ops.object.delete()
# Create Character A primitive
bpy.ops.mesh.primitive_capsule_add(radius=0.3, height=1.8, location=(0, 0, 0))
actor = bpy.context.active_object
actor.name = "Char_A"
# Set keyframes for walking
actor.location = (0, 0, 0)
actor.keyframe_insert(data_path="location", frame=1)
actor.location = (3, 4, 0)
actor.keyframe_insert(data_path="location", frame=144)
# Configure Camera
bpy.ops.object.camera_add(location=(-1, 0, 0.5), rotation=(1.57, 0, -0.6))
cam = bpy.context.active_object
bpy.context.scene.camera = cam
cam.data.lens = 24
# Render settings
bpy.context.scene.frame_start = 1
bpy.context.scene.frame_end = 150
bpy.context.scene.render.fps = 30
bpy.context.scene.render.image_settings.file_format = 'PNG'
if __name__ == "__main__":
setup_scene()
Next steps
- Run
blender --python buildPrevis.pyto generate the.blendfile. - Review the rendered MP4 to ensure the low-angle camera does not clip through the floor.
- Import the
previs.mp4into your H3 storyboard as a motion reference.
Connects securely to your tools. The creator never sees your data.
About this skill
This skill solves the problem of spatial and temporal uncertainty in 3D production. It eliminates guesswork by creating low-fidelity "graybox" previs in Blender to validate blocking, camera paths, and timing before committing to high-res rendering or complex character animation. The skill focuses on physical viability rather than visual polish, ensuring that character positions, camera trajectories, and set boundaries function correctly within the 3D space.
What it does
- Spatial Validation builds low-poly scene layouts using geometric primitives to verify character spacing, wall boundaries, and obstacles.
- Camera Orchestration configures focal lengths, camera paths, and look-at targets to confirm shots are physically achievable.
- Timing Synchronization maps audio duration to frame counts, creating motion references that align key events with the soundtrack timeline.
- Automated Rebuilds generates Python scripts to reconstruct the previs scene in any Blender environment without manual re-tweaking.
- Reference Integration formats previs data for H3 motion reference blocks, explicitly separating coarse motion from final visual style.
How it works
- Analyze Requirements by defining character relationships, spatial scales, and key action timestamps in a structured plan.
- Construct Graybox using spheres, cylinders, and cubes to represent actors and environment assets with flat color coding for identification.
- Animate Trajectories by setting keyframes for movement and camera motion, ensuring no clipping or boundary violations.
- Render and Encode outputting PNG sequences via Blender's Workbench or Eevee engines, then assembling into MP4 files using FFmpeg.
Frameworks & tools
The skill requires Blender (accessible via CLI) and FFmpeg for video encoding. It utilizes the bpy Python API for scene construction and script generation.
Why this beats prompting it yourself
Generic prompts often result in physically impossible camera moves or character placements that clip through walls. This skill enforces strict coordinate conventions, real-world camera constraints, and automated script generation, providing a repeatable technical pipeline instead of just a visual guess.
Use cases
- Validating if a complex tracking shot will hit a wall or obstacle.
- Checking if character blocking matches the duration of a specific audio line.
- Creating a motion guide for AI video generators to follow specific trajectories.
- Confirming microphone and prop placement relative to character height.
Known limitations
Does not support facial expressions, lip-syncing, or cloth simulation. Cannot be used for final character modeling or high-fidelity lighting. Requires Blender and FFmpeg installed on the host system.
No facial/lip-sync animation. No complex physics or cloth simulation. Requires local Blender and FFmpeg installations. Not for final visual styling.
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 19 days ago
Permissions
No special permissions declared or detected