Files

T

Adrian PerezandClaude Opus 4.7 f230d9a401 test(mcp): Villa Azul + 10-agent deep verification

Builds a larger, richer house than Casa del Sol via MCP save_scene
(no injection hack), then dispatches 10 parallel verifiers across
schema, geometry, dimensions, openings, HTTP, page render,
parentage, round-trip, spatial, visual.

Villa Azul — 56 nodes, validate_scene=true, 44KB on disk:
- 15x10m building envelope (vs Casa del Sol's 12x8)
- 9 interior zones (master bed/bath, bed 2/3, shared bath,
  living/dining, kitchen, entry hall, corridor)
- 10 doors + 12 windows (all cut successfully)
- 4 exterior zones (pool 8x4 + basin slab at -2m, outdoor kitchen,
  driveway, back patio)
- 5 rail-style fences (vs Casa del Sol's privacy) with 2m entrance gap

Verification: 108 checks, 104 PASS, 4 findings:
- V1 schema: 56/56
- V2 geometry: 7/7 (perimeter closes, interior T-junctions, no
  zone overlaps, fence gap verified)
- V3 dimensions: 13/13 zone areas exact (1 spec mismatch on site
  polygon default, not a build bug)
- V4 openings: 22/22 dimensional fit, surfaced a tool gap in
  cut_opening (no adjacency check) + my build packed too tightly
- V5 HTTP: 10/10 (GET/PUT/PATCH/DELETE/HEAD, If-Match conflicts)
- V6 page: 14/14 (/scene/:id 81KB, /scenes 20KB, 404 fallback)
- V7 parentage: surfaced CROSS_CUTTING §2 site->building->level
  parentId=null (pre-existing in core's loadScene)
- V8 round-trip: 10/10 byte-equal, duplicate_level -> 110 nodes
- V9 spatial: 12/12 (find_nodes, measure, constraints resource)
- V10 visual: HTML fallback (Chrome extension disconnected during
  run); API layer intact

Follow-up tracked: `cut_opening` should check opening-adjacency on
the same wall (minimum gap) to catch tight packing during patch
construction. Currently returns success and relies on the UI to
visualise the overlap.

Live at http://localhost:3002/scene/a6e7919eacbe.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

2026-04-19 20:39:08 +02:00

2.2 KiB

Raw Blame History

Villa Azul — Phase 9 V10 Visual Verification

Date: 2026-04-18
Scene: a6e7919eacbe ("Villa Azul")
Editor: http://localhost:3002/scene/a6e7919eacbe

Chrome MCP status: DISCONNECTED

mcp__claude-in-chrome__tabs_context_mcp returned "No such tool available". Per Phase 9 instructions, falling back to HTML / API diffing (Phase 8 P7 pattern). No screenshot IDs available.

HTTP probes

Target	Status	Bytes
`GET /scene/a6e7919eacbe`	200	81 737
`GET /scenes`	200	20 022
`GET /api/scenes`	200	Villa Azul present (id `a6e7919eacbe`, 56 nodes)
`GET /api/scenes/a6e7919eacbe`	200	23 625 (full graph)

Scenes-list HTML contains exactly the expected references: Villa Azul x2, a6e7919eacbe x3. No stray or duplicate entries — scene appears once in the browseable list (check 7 satisfied at the HTML level).

Graph inventory (from /api/scenes/:id)

Node-type census parsed from the returned JSON:

site: 1
building: 2 (the top-level site re-embeds its building child — expected)
level: 1
wall: 12
window: 12
door: 10
panel: 20 (door/window sub-children)
zone: 13
slab: 1 (slab_azul_pool — pool present)
fence: 5
polygon: 1 (site footprint)

Total 56 nodes, matches nodeCount. Hierarchy under level_r58jrtlaqqfx4rf0 lists 12 walls, 13 zones, 1 slab (pool), 5 fences — the sidebar tree would show site -> building -> level with these children (check 4 satisfied structurally).

What the scene "looks like" (inferred from geometry)

30x30 m site polygon, centred at origin.
12 exterior/interior walls, 2.8 m tall, 0.22 m thick, bounding a 15x10 m footprint (walls at y=-5 and y=5, x=-10 and x=5 plus interior partitions).
12 windows + 10 doors distributed across the walls.
13 zones (rooms/terraces).
Pool slab slab_azul_pool present as a dedicated node.
5 perimeter fences.

Console / runtime errors

Chrome disconnected, so read_console_messages not available. No HTTP 5xx observed; both editor and listing pages return 200.

Verdict

PASS (HTML fallback). Villa Azul exists, is listed once, loads a 56-node graph with the expected counts of walls (12), pool slab (1), and fences (5). Visual screenshot verification deferred — Chrome MCP not connected.

2.2 KiB Raw Blame History