stickman/docs/ideas/feature_backlog.md

Feature Backlog

This is the live list of future Stickman Studio feature ideas. Designer maintains it as ideas become clear in discussion with the user.

Suggested order

Priorities are recommendations for discussion, not a committed release schedule. Keep higher-priority items first and explain the reason for each ranking.

1. Move To with selectable movement styles — Status: Verified · Suggested priority: High. Intent: Give a stickman one destination action with a per-action style choice, starting with Walk (default) and Run; leave room for future styles like Fly. Use a fixed pace for each style initially, matched to its animation. Rationale: Children can direct the same route at different paces without choosing duplicate action types. The first run cycle should read as a believable human gait on the existing stickman rig, with clear foot contact and push-off, a flight phase, alternating arm swing, and cadence matched to travel speed. Connections: Phase 16 design brief, scripts/stage/action_registry.gd, scripts/editors/stage/action_editor.gd, scripts/stage/stage_scene_io.gd, scripts/characters/stickman/stickman_locomotion.gd, scripts/tools/create_animations.gd; existing scenes and rules currently store walk_to and should continue as Move To · Walk.

2. Distant background surfaces — Status: Planned · Suggested priority: High. Plan: Phase 19 — Background Ground and Wall Color. Intent: Add finite, inert Background ground panels behind buildings and walkable Ground, with textured Grass, Concrete, and colorable Carpet. Complete the Phase 19 wall Color flow for Paint, Brick, and Wood, with texture-preserving tint and eight shared recent colors. Rationale: Children can give a town or room a coherent setting while keeping the walkable route clear. The chooser and selected-piece inspector distinguish decorative scenery from solid Ground; backdrop and wall colors remain editable per piece. Horizon and edge treatments remain a separate idea. Connections: Phase 19 Scenery backlog item 5, scripts/stage/scenery/stage_surface_renderer.gd, scripts/stage/stage_scene_io.gd, scripts/editors/stage/sandbox_stage.gd.

3. Prop depth and behavior settings — Status: Verified · Suggested priority: Medium. Intent: Let placed props draw behind or in front of characters, and keep every prop selectable in EDIT even when covered. Dynamic sorting with characters is deferred until the Stage has a clear front/back movement convention. Rationale: Children can build layered sets with independently configurable visual depth, physics/support behavior, and interaction. Some props can provide surfaces that support other props while remaining pass-through for characters; others can block characters, move under physics, be picked up, or react to events. These roles should work across prop types rather than depend on one table-and-pot example. New placements start still and use a selected-prop settings card. Set Down targets a clicked stage-world release point, alongside free Drop and Throw; physics determines the subsequent landing on existing collision geometry, with no support-surface authoring. Door open/close behavior and scene travel stay in the later doorway feature. Phase 17a is implemented and user-verified. See the Phase 17 — Props for Storytelling plan and approved 17a UI specification. Connections: scripts/props/prop_block.gd, scripts/stage/stage_scene_io.gd, scripts/stage/stage_spawner.gd, scripts/stage/stage_selection.gd, scripts/editors/stage/sandbox_stage.gd, scripts/characters/stickman/stickman_prop_interactor.gd, Phase 11 prop interactions.

4. Functional doorways and doors — Status: Idea · Suggested priority: Medium. Intent: Let placed door artwork serve as scenery, an open passage, or a closed wall door that blocks characters until opened; an open passage may optionally link to another scene and arrival point. Rationale: Children can reuse door art for different stage roles. Door open/close actions and scene travel add state and depend on the basic prop blocking model and dynamic scene loading; keep this as later work. Whether a doorway activates automatically or through a directed action remains open. Connections: scripts/stage/stage_scene_io.gd, scripts/stage/trigger_area.gd, scripts/stage/trigger_registry.gd, scripts/editors/stage/sandbox_stage.gd.

5. Scenery, backgrounds, and vector surface finishes — Status: Planned · Suggested priority: Medium. Plan: Phase 19 — Scenery, Backgrounds, and Surface Finishes; Starter Scene Composition brief; composition UI/UX spec; composition architecture spec. Intent: Give the Stage a Scenery category with three distinct uses: a scene-wide Sky background, finite wall panels, and individually placed artwork. One chosen Sky preset fills the camera view; its gradient stays fixed to the screen while details respond gently and automatically to camera pan and zoom in EDIT, DIRECT, and PLAY; the initial designed presets are Clear Day, Cloudy Day, Sunset, and Starry Night. Skies are adaptive scene themes rather than .prop files: the .scene saves a stable preset ID, while each preset composes crisp vector gradients, varied clouds or stars, and sun or moon details without visible panel edges or obvious repetition. Wall panels are resizable, colorable rectangles with finishes such as paint, brick, and wood; .prop artwork can be placed as inert scenery. Scenery artwork supports simple Far and Near layers so a muted skyline strip can sit behind colorful storefronts, houses, and trees while all scenery remains behind the action. New Scene offers Blank and one polished Town Street starter built from a Cloudy Day Sky, Far skyline, Near storefront/tree scenery, and sidewalk/road Ground; the starter creates an unsaved scene whose ordinary parts remain editable. The focused starter-composition work also aligns Town Street and Living Room artwork to continuous, pan-safe ground/floor runs while preserving separate editable pieces. User-created scenery artwork reuses the existing vector artwork editor and .prop format; no separate Scenery Editor or .scenery format is needed initially. The Stage placement category, saved in the .scene, determines whether the same artwork is inert Scenery or an interactive Prop. Visual floor finishes are chosen through the existing Ground tool and remain walkable terrain; initial outdoor looks include concrete/sidewalk and asphalt/pavement alongside indoor finishes. Wall panels are inert Scenery. Shared finish names, previews, and color controls make the relationship clear without merging their behavior. Choosing a Ground finish affects new placement, while an existing piece changes only when directly edited. Ground alone gains finishes in this version; Ramp and Step retain their current appearance and remain unconstrained for a possible future terrain redesign. The artwork catalog uses optional multi-select tags from a built-in list: Nature, Buildings, Walls, Roads & Paths, and Objects. The same tags may filter both Props and Scenery because they describe the drawing, while the Stage entry point determines behavior. Rationale: Children can build outdoor and indoor sets without changing movement or interaction. A scene-wide Sky avoids accidental edges during camera movement, wall panels can build finite rooms over it, and cutout artwork dresses either setting. Curated long vector assets and the editable Town Street starter make a layered outdoor scene immediately achievable without adding a general template system. Reusing the established multi-shape vector authoring workflow avoids duplicate editors, split libraries, and file conversion. Its child-facing presentation may evolve from Prop Editor toward Artwork Editor so prop-only physics and attachment controls do not dominate scenery creation. Scenery objects and wall panels are explicitly inert: they cannot block, support, move, be picked up or used, or participate in events. Sky sits farthest back, wall panels and Far scenery render over it, Near scenery renders over those, and terrain, props, and actors remain in front. Ground patterns align across adjacent painted pieces so a floor or road reads as one continuous surface. A dedicated scenery-editing flow should keep broad or overlapping background art from blocking normal selection. Artwork may appear in several browse categories, while older untagged artwork remains available under All Artwork and Unsorted. Solid walls remain terrain or future static structures, while loose bricks and planks remain props. Individual clouds, suns, or moons may still be .prop scenery cutouts, but whole adaptive skies are not. Autonomous weather animation, multi-layer parallax scenery, directed sky changes, user-authored sky themes, custom repeatable finish authoring, a larger starter gallery, inserting starters into existing scenes, and user-authored set templates are future workflows and do not require a scenery asset type now. Connections: scripts/editors/prop/prop_editor.gd, scripts/editors/prop/prop_file.gd, scripts/props/prop_library.gd, scripts/editors/stage/stage_asset_browser.gd, scripts/terrain/terrain_block.gd, scripts/editors/stage/stage_terrain_painter.gd, scripts/stage/stage_scene_io.gd, scripts/editors/stage/stage_scene_adapter.gd, scripts/stage/stage_selection.gd, scripts/editors/stage/sandbox_stage.gd, Stage panel UI system.

   Sky scope: Choosing a Sky preset changes only the sky; it does not tint or recolor terrain, props, characters, or other scenery.

   Indoor layering: In addition to Far and Near backdrop scenery, an On the floor scenery placement lets inert rugs and floor markings render over finished Ground but beneath characters and interactive furniture. Ground remains in front of Far/Near scenery; floor decorations do not add collision or intercept ordinary object selection.

   Starter scope: New Scene offers Blank Scene, Town Street, and Living Room. Living Room combines a painted wall panel, wood Ground, Near wall decorations, an On the floor rug, and furniture Props; all ordinary parts remain editable in the new unsaved scene.

   Built-in content: The starters require a curated vector asset pack: four Sky themes; repeatable wall/Ground finishes; Town skyline, storefront, and tree scenery; Living Room window, fireplace, clock, shelf/books, pictures, plant, and rug scenery; and couch, armchair, and coffee-table Props. Raster concepts may guide the art, but shipped cutouts use repo-native vector/.prop assets and the Sky/finish visuals must remain crisp through Stage zoom.

   Art tracking: The accepted phase brief must include an asset production manifest with stable IDs, asset class/final format, tags/default layer, starter usage, behavior/collision expectation, source/rights, final location, visual acceptance, validation evidence, and status. Use Needed → Concept ready → Art approved → Final vector ready → Integrated → Verified so raster concepts cannot be mistaken for finished vector assets.

   Structural scope: Multi-level traversal and structural stairways are a separate future feature. House artwork, wall panels, and decorative stairs do not create reliable walkable floors or routes between elevations.

6. Multi-level walkable scenes — Status: Exploring · Suggested priority: Medium. Intent: Let children build a two-floor cutaway scene and direct a stickman with Move To from either floor to the other through one walkable stair connection. Rationale: This is separate follow-up work after the scenery feature, so navigation and locomotion changes do not delay visual set building. Upstairs/downstairs stories need reliable travel, correct-floor arrival, visible route feedback, and an understandable No path upstairs cue when disconnected; visual scenery alone cannot provide this. A narrow first slice uses editable lower and upper solid floors plus one child-facing Stairs connector, supports travel in both directions, and offers a way to frame both floors. The Stage has Ground, Ramp, and Step terrain with short-gap navigation links, but current movement and arrival are largely horizontal, so this needs a separate navigation and locomotion design. Keep Stairs independent from today's Ramp/Step taxonomy so their future redesign remains open. Defer additional floors, ladders, elevators, jumping routes, moving platforms, automatic camera direction, functional doors, and general building generation. Connections: scripts/editors/stage/sandbox_stage.gd, scripts/characters/stickman/stickman_locomotion.gd, scripts/stage/stage_spawner.gd, Move To backlog item 1, Scenery backlog item 5, Functional doorways backlog item 4.

7. Whole-scene visual filters — Status: Idea · Suggested priority: Low. Intent: Let children give the complete scene a unified look such as sepia or black-and-white. Rationale: A filter can signal a memory, old film, or mood across the set and characters together. It is distinct from Sky, which changes only the background. Explore it after the scenery system is established, with special attention to keeping characters, selection cues, and Stage controls readable. Connections: scripts/editors/stage/sandbox_stage.gd, scripts/stage/stage_scene_io.gd, scripts/editors/stage/stage_scene_adapter.gd, Scenery backlog item 5.

8. Use spots on props — Status: Verified · Suggested priority: Medium. Intent: Let children mark a handle, button, or lever on a placed prop and direct a stickman to use it. Rationale: A Use spot is distinct from a grip anchor: the prop stays in place while the stickman reaches to it. Initial direction is an instance-level spot that produces a Used prop event after the stickman reaches and completes the interaction. Rules can respond to use of the whole prop or filter for one specific spot; the event identifies the stickman, prop, and spot. Built-in effects such as toggling a light are deferred. The Stage needs a clear standing/reach target, visible EDIT feedback, and an understandable unreachable cue. Phase 17b is implemented and user-verified. See the Phase 17 — Props for Storytelling plan. Connections: scripts/props/prop_anchor_data.gd, scripts/props/prop_block.gd, scripts/characters/stickman/stickman_prop_interactor.gd, scripts/stage/action_registry.gd, scripts/stage/trigger_registry.gd.

9. Additional prop impact reactions — Status: Idea · Suggested priority: Medium. Intent: Explore toppling or breaking props when they are struck or thrown. Rationale: The wobble response in Phase 17c is implemented and user-verified; this idea covers further reactions such as toppling or breaking, which remain future options. The wobble implementation is documented in the Phase 17 plan. Connections: scripts/props/prop_block.gd, scripts/editors/stage/sandbox_stage.gd, scripts/stage/trigger_registry.gd.

10. Group props into set pieces — Status: Parked · Suggested priority: Low. Intent: Let children group objects for organization and moving together within one scene. Rationale: Deferred until later at the user's direction. Scene-only groups make arranging furnished sets faster. In PLAY, each item keeps its own physics and interaction; picking up one prop does not move the rest, and stopping PLAY returns the grouped items to their authored arrangement. Reusable set-piece assets are deferred. This is Phase 17d in the Phase 17 — Props for Storytelling plan. Connections: scripts/stage/stage_selection.gd, scripts/editors/stage/sandbox_stage.gd, scripts/stage/stage_scene_io.gd.

11. Stage-relative prop and scenery scale — Status: Planned — Phase 19 addition; UI approved · Suggested priority: Medium. Intent: Establish one recognizable Stage scale based on the standard stickman, then author each asset at the size its role calls for so ordinary placements start correctly without routine resizing. Rationale: The Prop Editor already shows a stickman guide, but its current Normalize Size target is 96 world units against an approximately 940-unit standing figure; current artwork dimensions also vary widely. The approved Phase 19 addition replaces Normalize Size with a reversible Size for Stage preview: Small/Room/Big starts plus fine adjustment, a foreground 1:1 figure, and a floor guide. Artist chooses ordinary dimensions from asset intent; new placements use authored size. Individual Scenery resize commits on release, while group resize keeps its existing Keep/Cancel preview. Loading performs no automatic normalization. Ordinary Save and live-reference behavior remain, with no new copy/version prompts. Phase 19 user/editor verification remains pending. Connections: Phase 10a Prop Scaling, Phase 19 Scenery, Backgrounds, and Surface Finishes, approved Phase 19 UI, scripts/editors/prop/prop_editor.gd, scripts/editors/prop/prop_canvas.gd, scripts/editors/stickman/whole_stickman_preview.gd.

12. Size-aware artwork previews — Status: Idea · Suggested priority: Low. Intent: Show a stickman silhouette or simple relative-size cue in artwork catalog previews so children can judge an asset before placing it. Rationale: Once asset dimensions are authored consistently, a size cue could reduce trial-and-error in the Stage without forcing objects to a fixed size. This depends on a settled Stage scale convention and should remain a browse-time hint, not an automatic resize. Connections: Stage-relative prop and scenery scale (backlog item 11), scripts/editors/stage/stage_asset_browser.gd, scripts/shared/ui/asset_card.gd.

13. Backdrop horizon and edge treatment — Status: Idea · Suggested priority: Medium. Intent: Make the transitions between Sky, broad background surfaces, scenery, and walkable Ground easy to understand and visually intentional. Rationale: A grass backdrop can look like a floating rectangle if its edge does not meet the scene composition cleanly. A horizon line, soft terrain contour, or deliberate overlap could help establish depth without adding collision. This remains separate from Phase 19's camera-aware Sky details. Connections: Distant background surfaces (backlog item 2), Scenery backlog item 5, Phase 19 — Background Ground and Wall Color.

14. Shared scenery color palette — Status: Idea · Suggested priority: Low. Intent: Reuse a small set of accessible color swatches across wall finishes and recolorable background materials. Rationale: Familiar choices help children coordinate walls, carpet, and other set pieces without opening a full picker for every edit. This should complement a full color picker where requested and preserve material texture. Connections: Distant background surfaces (backlog item 2), Scenery backlog item 5.

15. Duplicate background surface — Status: Idea · Suggested priority: Low. Intent: Let children duplicate a selected distant surface to extend a backdrop while keeping its material and color. Rationale: Repeating a configured panel may be faster than rebuilding a long landscape or room background from scratch. Connections: Distant background surfaces (backlog item 2), scripts/stage/stage_selection.gd.

16. Gentle camera-aware Sky details — Status: Planned · Suggested priority: Medium. Plan: Phase 19 — Scenery, Backgrounds, and Surface Finishes, camera-aware Sky addendum. Intent: Let clouds, sun, moon, and stars respond gently and automatically to Stage camera pan and zoom while the Sky gradient always fills the screen. Children choose a Sky and direct the scene without adjusting a motion control. Rationale: A small amount of camera response connects the Sky details to a dramatic shot while keeping them visually distant and avoiding a new child-facing setting. Each preset can tune its own response so details stay composed, clear of the main action, and inside the viewport. Background ground remains world-anchored and supplies the horizon. Keep the response bounded across EDIT, DIRECT, PLAY, viewport resize, and saved camera framing; do not add a slider or make Sky details establish a ground line. Connections: Scenery backlog item 5, Backdrop horizon and edge treatment (backlog item 13), scripts/stage/scenery/stage_sky_view.gd, scripts/stage/scenery/stage_background_ground.gd, scripts/editors/stage/sandbox_stage.gd.

Entry format

For each feature, record:

• Status: Idea, Exploring, Planned, Verified, or Parked
• Suggested priority: High, Medium, or Low
• Intent: The user goal or experience the feature should support
• Rationale: Why it may be valuable and how it fits the project
• Connections: Relevant project docs, code, and dependencies, when known

Historical context

The archived roadmap is historical material, not an active priority list. Revisit individual ideas from it only when they are relevant to a current discussion.