Color is a character's role tag, not their clothing. The previz does not convey wardrobe, expression or lighting (that belongs in the shot prompt); color only lets whoever reads the clip tell people apart — so it is one color per character per shot, a clash is swapped for an unused palette color, and the exported previz is handed to the video model with a line saying which color is whom.
The two dropdowns under the rig are what they are carrying: one item per hand (pistol, rifle, flashlight, knife, baton, radio, phone, plus cup, bag, umbrella, book, tray, microphone, sword). Each item knows which hand it belongs in, so "pistol in one hand, flashlight in the other" is two picks and nothing else. Every rig holds things: the realistic rig in its hand bones, the stand-in across the chest, the simple rig in the palm.
Posing has two layers. Start from Pose presets to get roughly there, then open Pose controls and turn each joint into place: body, torso, head, left and right shoulder, elbow, hip and knee, one slider each, in degrees. The presets come in two batches: ones that strike an attitude (Stand, Walk, Run, Sit, Crouch, Kneel, Hands on hips, Bow, Think, Wave, Arms crossed, Checking phone…) everyday actions (Point, Open door, Carry, Handshake, Offer, Look back, Hand to mouth, Seated typing, Read, Lie down, Hands over face, Weapon ready), and police-action ones (Aim, Low ready, Fire, Crouch aim, Kick door, Take cover, Drag, Hostage hold, Hands up, Hit, Cuff, Sweep).