Ready to use Six full-body action poses in one 3x2 grid
Use this skillSame 3x2 landscape grid as Doodle Collage, but every panel shows the full body performing a different dynamic action — dancing, jumping, walking, stretching — with motion-line doodle overlays that trace the movement across the page.
Use when the user asks for full-body, head-to-toe, or action poses from one photo — for example "dancing", "karate", "jumping", "action shot", "full body", or "my whole body doing stuff".
Every skill is one SKILL.md — this is the exact file the agent reads.
Turn one uploaded photo into a single wide landscape image laid out as a strict 3-column by 2-row grid, where every panel shows the character head to toe in motion.
Pick this skill when the request involves movement or the whole body: dancing, jumping, running, karate, yoga, “action shot”, “full body”, “head to toe”, “show my outfit”.
Choose a different skill instead when the user asks for:
doodle-avatardoodle-collageSix panels, same illustrated character in all of them, each performing a
different dynamic action. The critical constraint: every panel shows the
complete figure from head to feet — never cropped to the face or torso.
That is the only thing separating this skill from doodle-collage.
Because the whole body is visible, the outfit matters as much as the face. Carry the clothing, colors, and silhouette across from the photo, and keep them identical in all six panels.
Motion-line doodle overlays trace the movement across the page. See
references/pose-library.md for action ideas,
../doodle-collage/references/grid-layout.md for the grid spec, and
../doodle-avatar/references/style-guide.md for the house style.
doodle-collage as the
better fit before proceeding.generateDoodle with skill: "full-body" and the uploaded photo
as imageUrl.Reference image: <url>, pass it as
refImageUrl.Offer a refinement such as different actions, a specific sport or dance
style, or a switch to close-up panels with doodle-collage.
