Earthwalk / Art direction & acceptance

The ISS, down to
the difficult angles.

A reference-driven station rebuild. The suit you approved, brought to life. A sharper Earth with drifting clouds and a station that truly orbits it.

40station review packages
400station visual checks
60additional acceptance checks
0checks pre-scored as passing

APPROVED · 5 SEPTEMBER 2026
Implementation and reference comparison are underway. Acceptance results remain separate from these requirements.

The standard before implementation

Correct structure comes first

Real dimensions, correct module connections, accurate silhouettes, and one dated configuration. Fine surface detail comes after the shape and layout pass.

No average can hide a bad part

Every check needs evidence and a score of at least 9/10. Wrong placement, clipping, missing hardware, or an unsupported quality claim blocks completion.

The target

Build a convincing, one-to-one ISS at metre scale, beautiful in a full-station view and credible when an astronaut approaches a joint, window, rail, solar wing, or radiator. Keep the astronaut appearance you approved and add natural limb motion. Make Earth sharper, its clouds move slowly, and the station actually orbit it.

This is the approved acceptance rubric, dated 5 September 2026. It contains 40 station rubrics with 10 distinct visual checks each: 400 station checks. Six additional ten-check gates cover animation, controls, Earth, orbit, interiors, and scene-wide acceptance: 461 checks in total including one branding criterion assessed on both the suit and ISS. These are approved requirements. Current assessments and unresolved defects are recorded separately in the evidence ledger and review report.

Approved by the user on 5 September 2026. Implementation is underway. The OpenAI logo requirement is added as B.01, without replacing any of the 400 station checks.

What one-to-one means here

Approved configuration: the permanent station immediately after the 15 June 2023 spacewalk, with six installed iROSA wings. Before modeling, freeze a configuration ledger with a reference timestamp, module connections, transferred hardware, external payloads, visiting vehicles, and articulated poses. Check the ledger against dated photographs; never mix incompatible installation eras. The exact pose and visiting-vehicle manifest are still to be researched, not silently assumed.

Every visible station assembly must match its real shape, size, position, orientation, construction, and material as far as the evidence establishes. Use published dimensions for metre-scale geometry and matched photographs for local features. A convincing generic module cannot pass as the real module.

Forty is the number of review packages, not a claim that the ISS has only forty physical parts. Repeated wings, platforms, adapters, radiators, and payloads receive separate child-instance records. Their lowest result controls the package; they are not averaged together. Antennas, tanks, plumbing, and small equipment belong to their actual host assembly. Visiting vehicles receive child rubrics using the same ten-category evidence method, beyond the base 400, once the manifest is fixed.

The approved Armstrong suit is intentionally retained as a historical avatar; it is not relabeled as a period-correct ISS EMU. View-button transitions between inside and outside remain acceptable. Door animation is optional. All player-accessible interiors must follow the actual module arrangement.

Unknown or obscured details remain marked Unverified. A plausible reconstruction may be used in an interim candidate, but it cannot be described as a verified one-to-one match. If sources conflict, resolve their dates and provenance before changing the model.

Configuration anchor: NASA's 15 June 2023 iROSA completion report.

Evidence required for every part

Create a reference board with source URL, agency/creator, image identifier, capture date, usage terms, relevant crop, dimensions, and confidence. Begin with the official reference anchors linked below, then collect part-specific photographs. Those anchors are not yet 400 completed reference comparisons.

Capture forward, aft, port, starboard, zenith, and nadir views, plus two three-quarter views, a critical junction close-up, and an underside/back-side close-up: at least ten rendered views per package. For repeated hardware, repeat the affected views per instance. Add a slow full orbit and close flyby. Capture first at the nearest permitted camera distance, then at roughly 5–15 m and station-wide distances, adjusted to the assembly size.

For comparison, match camera direction, perspective, framing, configuration, appendage pose, and lighting. Use side-by-side crops and silhouette/landmark overlays. Do not distort the render to force an overlay fit or use screenshots with unmatched exposure as proof of material accuracy. At least three distinct source angles per assembly are the collection target; if coverage is unavailable, name the unverified surface.

Use shaded beauty, neutral-light, wireframe, normal, and material-isolation views to diagnose faults. These diagnostic passes support the final native-resolution beauty and motion captures; they cannot replace them. Keep the Sun, camera exposure, and color handling recorded for reproducibility.

Every check gets an evidence row: check ID, child instance, build ID, source image IDs, camera/lighting preset, render or clip, measured result, score, status, defect, and retest result. No screenshots from an older candidate can prove a changed candidate passes.

Scoring and rejection rules

Score each check from 0–10, with Unverified kept separate from zero. 0 = missing; 1–4 = substantially wrong or broken; 5 = functional prototype with obvious approximation; 6–7 = recognizable but visibly inaccurate; 8 = close with an identifiable mismatch; 9 = reference-supported match with only minor, documented uncertainty and no visible defect in the tested views; 10 = no reproducible discrepancy within the documented source precision and viewing coverage. Ten is not a claim of absolute perfection.

Acceptance: every required check must score at least 9 and have its evidence attached. Each package reports its minimum score, not an average. Wrong module placement, missing required hardware, visible clipping, exposed mesh intersections, floating supports, broken animation, or unresolved configuration mismatches are hard failures regardless of score.

Dimensional tolerance: for a reliably dimensioned major body or assembly, error must be no more than the larger of 0.5% and the published measurement uncertainty. For matched image landmarks, target a median error at most 1% and a maximum at most 2% of the assembly image bounding-box diagonal. These are project QA thresholds, not NASA tolerances, and cannot excuse a visibly wrong feature. If camera calibration or reference precision is inadequate, mark the measurement Unverified.

Configuration identity, port connections, assembly counts, and mirrored orientation require exact agreement with the frozen ledger. Exposed cracks, unintended overlaps, flicker, missing faces, and clipping in reachable views have zero allowed reproducible occurrences. Intentional mechanical nesting and hidden construction overlaps are acceptable only when they produce the referenced physical appearance.

Every new asset must pass the same tests. Inspect mesh quality, scale, normals, UVs, texture resolution, material response, license, and configuration before accepting it. Real-world provenance is useful evidence, not a quality exemption.

A check may be marked Not applicable only with a written reason that the feature truly does not exist in that instance. Missing references, difficult modeling, or performance cost do not justify N/A. Replacement checks must preserve the ten-check coverage. Do not silently lower thresholds to make a candidate pass.

How iteration will work after review

1. Freeze the current suit and working controls as a comparison baseline. Build the dated station inventory and reference boards. Resolve module layout, dimensions, attachment axes, visiting vehicles, and appendage poses before detailed art work.

2. Reconstruct the station at the correct scale using simple diagnostic geometry. Compare all major silhouettes and connection points. Reject incorrect proportions or layout now; surface detail cannot repair them.

3. Work through the 40 packages: structure and joints, secondary equipment, materials, then close-up details. After each pass, capture the prescribed views, enter defects, correct them, and repeat the affected comparisons. Recheck neighboring parts whenever an interface changes.

4. Rig and animate the approved suit, implement independent camera orbit/body control, improve Earth, and add orbital motion. Evaluate their separate gates and recheck station contacts, lighting, camera behavior, and performance.

5. Run isolated Playwright self-play with repeatable inputs, native screenshots, motion clips, console inspection, and named collision routes. Review the visible results yourself in the captured frames; automated assertions do not establish visual fidelity.

6. Perform a fresh whole-scene review against the full ledger, including back sides and difficult angles. Keep dated before/after captures and rejected candidates. Repeat while required defects remain; neither a green test suite nor an arbitrary iteration count is a stopping rule.

7. Finish only when the exact candidate passes the complete ledger and the final evidence package is reviewable. If evidence, asset availability, or hardware limits prevent a requirement from passing, report that specific gap and keep the fidelity claim bounded; do not rename incomplete work as complete.

40 parts. Ten visual checks each.

Each card contains specific criteria for that assembly. Every repeated instance must pass independently. Expand a card, search for a feature, or open everything for a full read.

40 of 40 station parts · 400 criteria · Results in the evidence ledger
01Unity · Node 1Modules · 10 checks · Unreviewed

Shell, port collars, exposed fittings, and the visible vestibules; connected modules have their own rubrics.

Reference anchors: R1. Matched photo boards and scores are pending.

  1. S01.01Match the measured cylinder length, diameter, end profiles, and port-center spacing in orthographic overlays.
  2. S01.02Verify every occupied and unused port against the frozen configuration; a plausible-looking but wrong connection fails.
  3. S01.03Match axial and radial collar depth, flange steps, hatch recesses, and insulation transitions in oblique views.
  4. S01.04Reconstruct the visible shielding-panel layout; reject uniformly repeated panels where the reference is asymmetric.
  5. S01.05Place exterior rails and their standoffs from photographs; check clearance between rails and the pressure-shell covering.
  6. S01.06Resolve exposed cable exits, connectors, and brackets as attached hardware, without wires vanishing into arbitrary surfaces.
  7. S01.07Match white shielding, blanket seams, and exposed metal separately under calibrated sunlight and Earth fill.
  8. S01.08Compare forward, aft, port, starboard, zenith, and nadir views; adjacent modules must neither hide errors nor intersect Unity.
  9. S01.09Inspect port interiors from the accessible corridor: wall thickness, hatch-frame depth, and structural seams remain continuous.
  10. S01.10Sweep a close camera around the module; reject faceted curves, flipped triangles, texture swimming, and flashing seams.
02Harmony · Node 2Modules · 10 checks · Unreviewed

Node body and attachment interfaces; docking adapters are reviewed in part 22.

Reference anchors: R1. Matched photo boards and scores are pending.

  1. S02.01Match Harmony's length-to-diameter ratio and end taper to dimensional and photographic references.
  2. S02.02Verify its actual connection to Destiny and the lateral positions of Columbus and Kibo; no mirrored layout.
  3. S02.03Check the forward and zenith docking-interface positions and orientation against the selected configuration.
  4. S02.04Compare each visible shielding strip and collar interruption rather than duplicating Unity's surface pattern.
  5. S02.05Match handrail lengths, gaps, bracket feet, and access paths around protruding equipment.
  6. S02.06Place grapple and utility fittings at referenced mounting points, with their bases seated on the module.
  7. S02.07Inspect docking-adapter junctions from underneath for floating flanges, shell penetration, and incorrect gasket depth.
  8. S02.08Match blanket texture scale and metal roughness without embossing the entire body with oversized fabric weave.
  9. S02.09Compare the node's silhouette with both laboratories attached from front, top, bottom, and rear.
  10. S02.10Inspect a moving close-up through the node and around its exterior; panel joins and collar shading stay stable.
03Tranquility · Node 3Modules · 10 checks · Unreviewed

Node shell, ports, and visible interior junctions; its attached modules are separately scored.

Reference anchors: R1 · R4. Matched photo boards and scores are pending.

  1. S03.01Match the real node proportions and its orientation relative to Unity, with dimensional evidence.
  2. S03.02Map Cupola, Leonardo, BEAM, and Bishop to their correct dated ports; no arbitrary radial arrangement.
  3. S03.03Check occupied and capped port collar shapes from below and both ends of the node.
  4. S03.04Match shielding segmentation around the closely packed attachments; panel seams must terminate at real boundaries.
  5. S03.05Reproduce observed rails, brackets, and cable runs with a visible physical attachment at each end.
  6. S03.06Check clearance between adjacent attached modules and the node's exterior hardware from the narrowest viewing gaps.
  7. S03.07Match blanket folds, hard shielding, and rim metal as different surfaces under the same light.
  8. S03.08Inspect the Cupola junction from outside and inside; the viewing direction and passage alignment must agree.
  9. S03.09Compare the node and its attachments in a matched nadir view, preserving the reference's crowded but purposeful silhouette.
  10. S03.10Orbit the camera through the entire cluster; reject disappearing walls, overlapping skins, and unsupported equipment.
04Destiny laboratoryModules · 10 checks · Unreviewed

Laboratory shell, observation-window hardware, external interfaces, and visible laboratory interior.

Reference anchors: R1 · R14. Matched photo boards and scores are pending.

  1. S04.01Match the laboratory's long cylindrical profile, end geometry, and measured scale relative to an astronaut.
  2. S04.02Verify its axial connections to Unity and Harmony and its relationship to the central truss support.
  3. S04.03Match longitudinal shielding rows, circumferential breaks, and end-cap panel shapes in reference overlays.
  4. S04.04Place the observation-window opening and protective hardware on the correct side and at the correct longitudinal position.
  5. S04.05Inspect the window recess, rim thickness, glazing, and frame contact from inside and outside.
  6. S04.06Match external rails, fixtures, and support interfaces; reject evenly spaced decorative hardware without references.
  7. S04.07Check that the S0 support attachments meet the correct shell locations without visually penetrating the laboratory.
  8. S04.08Match laboratory rack spacing and passage dimensions in the explorable portion using interior photographs.
  9. S04.09Compare white shielding, glass, exposed metal, and interior textiles without flattening all materials into one finish.
  10. S04.10Traverse the corridor and orbit the exterior at close range; window edges, rack faces, and shell seams remain stable.
05Columbus laboratoryModules · 10 checks · Unreviewed

Laboratory body, mounting hardware, and dated external payload interfaces.

Reference anchors: R1 · R4. Matched photo boards and scores are pending.

  1. S05.01Match Columbus's compact proportions, end caps, and docking neck against module photographs and dimensions.
  2. S05.02Verify its side of Harmony and its axial orientation; a mirrored laboratory fails even if the silhouette looks attractive.
  3. S05.03Reproduce the module-specific exterior panel divisions and their seams at the end caps.
  4. S05.04Match the visible external payload mounting structures, including thickness and support feet.
  5. S05.05Inventory payloads against the chosen date; do not copy a launch-era installation into the modern configuration unchecked.
  6. S05.06Inspect handrails, labels, utility fittings, and their positions relative to the shielding boundaries.
  7. S05.07Compare the attachment neck and underside from oblique views for collars passing through the parent module.
  8. S05.08Match insulation and metal reflectance with the reference exposure normalized; avoid a uniformly chrome or plastic body.
  9. S05.09Check the visible interior rack arrangement and hatch direction against Columbus interior imagery.
  10. S05.10Sweep from a station-wide view to a one-metre approach; silhouette, fine detail, and texture sharpness remain coherent.
06Kibo pressurized moduleModules · 10 checks · Unreviewed

Main Japanese laboratory, end hardware, module airlock exterior, and visible interior.

Reference anchors: R3. Matched photo boards and scores are pending.

  1. S06.01Match the main laboratory's dimensions and the relative sizes of its cylindrical body and end structures.
  2. S06.02Verify the connection to Harmony and the placement of the logistics module and exposed facility around it.
  3. S06.03Match shielding strips, end-cap segmentation, and characteristic interruptions to the cylindrical silhouette.
  4. S06.04Reconstruct the experiment-airlock opening and adjacent structure at the photographed position and scale.
  5. S06.05Check the robotic-arm base interface and nearby fittings against JAXA imagery, without moving them for convenience.
  6. S06.06Match rail runs and external utility routing around the end of the laboratory.
  7. S06.07Inspect the exposed-facility junction for real brackets, correct spacing, and no intersecting shell surfaces.
  8. S06.08Compare white panels, blanket-covered fittings, window glass, and bare metal under one lighting setup.
  9. S06.09Match interior passage width, rack faces, and window/airlock relationships in the accessible area.
  10. S06.10Compare front, rear, underside, and elevated approach footage; no major feature may disappear at a detail transition.
07Kibo logistics module · ELM-PSModules · 10 checks · Unreviewed

Pressurized logistics section and its attachment; obsolete exposed logistics hardware must not be added from old diagrams.

Reference anchors: R3. Matched photo boards and scores are pending.

  1. S07.01Match the short cylinder's proportions, end contour, and diameter relative to the main Kibo laboratory.
  2. S07.02Verify the attachment location, axis, and rotational orientation against the frozen configuration.
  3. S07.03Compare the coupling neck and ring stack from multiple sides for correct length and flange depth.
  4. S07.04Match exterior panel and blanket boundaries without recycling the main laboratory's texture arrangement.
  5. S07.05Place the photographed grapple fitting and external access hardware with correctly scaled brackets.
  6. S07.06Check handrail spacing and projection from the skin in close-up side views.
  7. S07.07Inspect the lower junction and nearby structures for contact errors hidden in a frontal photograph.
  8. S07.08Match seams and blanket wrinkles at a realistic scale; reject coarse noise used to simulate detail.
  9. S07.09Verify visible identification markings against source orientation and prohibit mirrored lettering.
  10. S07.10Perform a full close orbit of the module; curved highlights, end-cap triangulation, and joints must remain clean.
08Kibo exposed facilityExternal science · 10 checks · Unreviewed

Facility platform, attached experiment units and communications hardware present in the selected configuration.

Reference anchors: R3. Matched photo boards and scores are pending.

  1. S08.01Match the platform outline, thickness, and supporting geometry to top, side, and underside references.
  2. S08.02Verify its location and alignment at the end of the pressurized module.
  3. S08.03Map each payload site to the dated inventory; empty and occupied positions must match the reference.
  4. S08.04Reconstruct the visible shapes of each installed experiment rather than filling sites with generic white boxes.
  5. S08.05Match payload attachment mechanisms, support brackets, and interfaces at close range.
  6. S08.06Check exposed cable and utility routing across the platform for supported paths and credible terminations.
  7. S08.07Inspect underside equipment and structure; the platform must not become a featureless slab from below.
  8. S08.08Match blanket, structural metal, experiment covers, and dark apertures as distinct materials.
  9. S08.09Compare payload heights and gaps in a low grazing view; reject intersecting instruments or floating mounting feet.
  10. S08.10Fly around all platform edges and beneath it; thin parts and instrument silhouettes must remain stable in motion.
09Kibo robotic arm · JEMRMSRobotics · 10 checks · Unreviewed

Main arm and any smaller arm/tool present in the frozen pose, including their mounts.

Reference anchors: R3. Matched photo boards and scores are pending.

  1. S09.01Match arm-segment lengths and joint housing sizes to JAXA dimensional and photographic references.
  2. S09.02Verify the base position and axis on Kibo, including the complete mounting connection.
  3. S09.03Match the documented reference pose; do not invent a convenient bend that intersects station equipment.
  4. S09.04Inspect each joint's axis, overlapping mechanical covers, and gap so the arm reads as a connected mechanism.
  5. S09.05Reconstruct the end effector, camera housings, and visible tool fittings instead of a rounded stump.
  6. S09.06Match blanket seams and exposed joint metal with plausible local texture scale.
  7. S09.07Route visible harnesses around joints without stretching across open gaps or entering solid housings incorrectly.
  8. S09.08Check the arm from front, rear, underside, and end-on for collapsed geometry and broken attachment continuity.
  9. S09.09Compare the arm's size and clearance above the exposed facility in a matched wide reference view.
  10. S09.10Inspect close moving footage through the joint chain; fine fittings must not flicker or merge into nearby payloads.
10CupolaModules · 10 checks · Unreviewed

Exterior dome, glazing, shutters, interior window frames, and visible observation workstation.

Reference anchors: R1 · R4. Matched photo boards and scores are pending.

  1. S10.01Match the dome proportions and the central and surrounding window arrangement to measured reference views.
  2. S10.02Verify its Earth-facing mounting direction and the exact junction to Tranquility.
  3. S10.03Match the shape, thickness, bevels, and alignment of every window frame, inside and outside.
  4. S10.04Model each shutter with visible hinges and the chosen open/closed pose; no unsupported floating covers.
  5. S10.05Check glazing thickness, reflections, transmission, and tint against photographs without obscuring Earth's view.
  6. S10.06Inspect interior-to-exterior continuity at each pane; no double frames, wall leaks, or mismatched window openings.
  7. S10.07Reproduce the visible interior equipment, handholds, and cable placement with a clear observation space.
  8. S10.08Compare real wear, insulation, and frame materials in controlled daylight rather than adding arbitrary grime.
  9. S10.09Match the Earth horizon and station obstruction through each window from a physically plausible viewing position.
  10. S10.10Move the camera across window edges and outside the dome; reject glass sorting errors, clipping, and unstable reflections.
11Quest airlockModules · 10 checks · Unreviewed

Equipment lock, crew lock, external tanks, exit frame, and visible interior. Door animation is optional.

Reference anchors: R1. Matched photo boards and scores are pending.

  1. S11.01Match the two connected chamber proportions, diameter change, and end contours against references.
  2. S11.02Verify the airlock's mounting point and orientation relative to Unity.
  3. S11.03Match the exterior gas-tank count, shapes, locations, supports, and straps to the selected configuration.
  4. S11.04Reconstruct the EVA exit opening and rim with real depth and correctly positioned neighboring handrails.
  5. S11.05Match external shielding, insulation seams, hatch markings, and access fittings to close reference photographs.
  6. S11.06Inspect the chamber junction and tank brackets from below for floating supports and exposed intersections.
  7. S11.07Reproduce the visible crew-lock interior volume, handholds, and suit-clearance envelope at actual scale.
  8. S11.08Match painted shielding, metal tank hardware, seals, and textile surfaces without identical roughness.
  9. S11.09Compare the full EVA approach and exit composition from both sides of the frame.
  10. S11.10Inspect the astronaut passing through a clear exit or using the view transition; no suit, tether, or camera penetration.
12Leonardo permanent multipurpose moduleModules · 10 checks · Unreviewed

Cargo module shell, its present attachment, and any visible storage interior.

Reference anchors: R1. Matched photo boards and scores are pending.

  1. S12.01Match the module's length, diameter, and end-cap profile to photographs and dimensions.
  2. S12.02Verify its June 2023 attachment location; launch and earlier berthing photographs are not location evidence.
  3. S12.03Match collar depth and the offset between the main body and attached node.
  4. S12.04Reproduce shielding bands and module-specific panel seams, including the end-cap transitions.
  5. S12.05Check exterior rails, grapple fittings, and bracket feet against source positions.
  6. S12.06Match visible labels and blanket closures without repeating generic markings around the cylinder.
  7. S12.07Inspect the crowded junction with Tranquility and neighboring modules from the underside.
  8. S12.08Compare shielding, insulation, and exposed rim metal with nearby modules under identical lighting.
  9. S12.09If visible inside, match cargo bag dimensions, restraint locations, and usable passage clearance to reference imagery.
  10. S12.10Sweep all accessible exterior sides and the visible interior; no floating bags, shell clipping, or unstable panel edges.
13BEAM expandable moduleModules · 10 checks · Unreviewed

Expanded envelope, rigid ends, mounting collar, and visible external hardware.

Reference anchors: R1. Matched photo boards and scores are pending.

  1. S13.01Match the expanded envelope's measured length, maximum width, and end-to-middle profile.
  2. S13.02Verify its actual mounting port and orientation on Tranquility.
  3. S13.03Match the fabric envelope's broad contour and seam placement without inflating it into a perfectly smooth balloon.
  4. S13.04Reproduce the transition from flexible envelope to rigid end structures with correct thickness and contact.
  5. S13.05Match the collar stack and attachment depth from side and underside photographs.
  6. S13.06Place observed straps, seams, access fittings, and markings on their referenced parts of the envelope.
  7. S13.07Compare soft blanket roughness with adjacent hard shielding under the same sunlight.
  8. S13.08Check silhouette shading for pinching triangles, lumpy tessellation, and exaggerated normal-map wrinkles.
  9. S13.09Inspect clearance to nearby modules, rails, and the Cupola cluster from the narrowest angles.
  10. S13.10Orbit close to both ends and along the envelope; no seam crawling, texture stretching, or detached end caps.
14Bishop airlockModules · 10 checks · Unreviewed

Airlock shell, berth interface, external mounting sites, and payloads present in the dated scene.

Reference anchors: R1. Matched photo boards and scores are pending.

  1. S14.01Match the distinctive shell outline and dimensions rather than substituting a generic cylinder.
  2. S14.02Verify its berth and orientation on Tranquility for the chosen reference date.
  3. S14.03Match the opening rim, attachment collar, and rear geometry from front and side photographs.
  4. S14.04Reconstruct exterior shielding facets and seams with clean, intentional panel intersections.
  5. S14.05Locate grapple fittings and their mounting plates at the correct scale and orientation.
  6. S14.06Inventory visible payload brackets and installed hardware against dated photographs.
  7. S14.07Inspect the underside and mounting junction for unsupported brackets and collars penetrating the node.
  8. S14.08Match blanket-covered areas, hard panels, and exposed metal with appropriate separate materials.
  9. S14.09Check its silhouette against BEAM, Cupola, and Leonardo in the same wide reference view.
  10. S14.10Fly around the airlock at close range; rims, fine mounts, and panel joins stay stable without hiding defects in shadow.
15ZaryaModules · 10 checks · Unreviewed

Complete module, its solar-array configuration, external fittings, and relevant docking interfaces.

Reference anchors: R1. Matched photo boards and scores are pending.

  1. S15.01Match the stepped body, tapered sections, end structures, and measured proportions to dated photographs.
  2. S15.02Verify its orientation and connections to the U.S. and Russian station segments.
  3. S15.03Match the actual solar-array deployment state in June 2023; older photographs cannot silently determine the pose.
  4. S15.04Reconstruct panel blankets and thermal-cover boundaries around each change in body diameter.
  5. S15.05Locate external tanks, conduits, antennas, and brackets from an itemized host-module inventory.
  6. S15.06Check docking-collar and adapter geometry from end-on and oblique views.
  7. S15.07Inspect array roots, hinges, and any folded structure for connected, nonintersecting geometry.
  8. S15.08Match insulation coloration and exposed hardware finish without painting the whole module one uniform color.
  9. S15.09Compare top, underside, and both side silhouettes, including asymmetric external equipment.
  10. S15.10Fly along the entire module and around its ends; cable routes, seams, and thin appendages remain clean in motion.
16Zvezda service moduleModules · 10 checks · Unreviewed

Service-module body, arrays, antennas, windows, docking end, and exterior equipment.

Reference anchors: R1. Matched photo boards and scores are pending.

  1. S16.01Match each stepped body section, tapered transition, and docking-end proportion to measured references.
  2. S16.02Verify its axis, connection to Zarya, and spatial relationship to attached Russian modules.
  3. S16.03Match both solar wings, their cell patterns, root geometry, and documented deployment orientation.
  4. S16.04Place visible windows and their frame/recess geometry accurately along the body.
  5. S16.05Inventory antenna shapes, booms, dishes, and mounting points against dated photographs.
  6. S16.06Match thermal blankets, access-panel boundaries, rails, and cable routes around the asymmetric exterior.
  7. S16.07Inspect the aft docking region and external thruster hardware in close reference views without inventing nozzle detail.
  8. S16.08Compare blanket reflectance, array coloration, glazing, and exposed metal under consistent lighting.
  9. S16.09Check all module and array clearances from top, underside, front, and rear.
  10. S16.10Orbit the full service module; no faceted tapers, disconnected booms, texture repetition, or flashing array edges.
17PoiskModules · 10 checks · Unreviewed

Mini-research module, docking end, access hardware, and its station attachment.

Reference anchors: R1. Matched photo boards and scores are pending.

  1. S17.01Match the compact body's contour, dimensional proportions, and end profile against Poisk-specific references.
  2. S17.02Verify the correct attachment point and outward direction; do not substitute Pirs geometry or location.
  3. S17.03Match the docking-end rim, recesses, and attachment depth from axial and side views.
  4. S17.04Reconstruct the exterior hatch shape and its position relative to the body seams.
  5. S17.05Place handrails, foot restraints, and brackets at photographed locations with correct standoff distances.
  6. S17.06Match blanket seams, straps, labels, and access-cover edges without generic repeated details.
  7. S17.07Check attached antenna and utility hardware against the dated inventory for this module.
  8. S17.08Compare the module's material response and weathering with adjacent Russian-segment surfaces.
  9. S17.09Inspect the root connection and every exterior fitting from underneath for floating or intersecting geometry.
  10. S17.10Fly around the entire contour at astronaut distance; hatch rims, rounded sections, and small brackets remain stable.
18RassvetModules · 10 checks · Unreviewed

Mini-research module and its dated external equipment; transferred hardware must be removed from obsolete positions.

Reference anchors: R1. Matched photo boards and scores are pending.

  1. S18.01Match the module's body proportions, narrowed sections, and docking-end geometry to reference dimensions.
  2. S18.02Verify the mounting point and orientation relative to Zarya and the rest of the Russian segment.
  3. S18.03Map external equipment to June 2023 after relevant transfers; reject copying an earlier cargo configuration.
  4. S18.04Match blanket panel shapes and seam locations around the distinctive body transitions.
  5. S18.05Reconstruct the visible docking rim, mechanisms, and surrounding hardware at credible scale.
  6. S18.06Place handrails and access fixtures with their exact observed spacing and standoff geometry.
  7. S18.07Route visible cables and hoses to real fittings rather than allowing them to disappear into blank surfaces.
  8. S18.08Compare insulation, metal, and dark recesses with exposure-matched close-up photographs.
  9. S18.09Inspect the crowded mounting junction from several oblique views for unintended overlaps.
  10. S18.10Perform an end-to-end close flyby and full orbit; small fittings, blanket detail, and collar seams stay visually coherent.
19Nauka multipurpose laboratoryModules · 10 checks · Unreviewed

Main body, arrays, transferred experiment airlock and radiator, exterior hardware; ERA is part 21.

Reference anchors: R1 · R5. Matched photo boards and scores are pending.

  1. S19.01Match the long stepped body, tapered transitions, and end structures to measured Nauka references.
  2. S19.02Verify the station connection and the position of Prichal relative to the main laboratory.
  3. S19.03Match solar-wing sizes, cells, hinges, and the deployment pose selected for the dated scene.
  4. S19.04Verify the transferred radiator and experiment-airlock locations and deployment state for June 2023.
  5. S19.05Reconstruct the ERA base points and their supporting fittings without duplicating the robot itself.
  6. S19.06Match exterior blankets, rails, hatch outlines, windows, and identifiable equipment placement.
  7. S19.07Inspect the underside, radiator mounting, and airlock junction for missing supports or exposed mesh penetration.
  8. S19.08Match the contrast among insulation, photovoltaic surfaces, radiator faces, and bare mechanisms.
  9. S19.09Compare the full module silhouette with its appendages from front, rear, nadir, and zenith.
  10. S19.10Fly through the reachable gaps around its appendages; geometry, materials, and contact points remain stable in motion.
20Prichal docking nodeModules · 10 checks · Unreviewed

Node body, all docking-port interfaces, caps, blankets, and Nauka connection.

Reference anchors: R1. Matched photo boards and scores are pending.

  1. S20.01Match the node's rounded body dimensions and its relationship to the diameter of each port.
  2. S20.02Verify the connection to Nauka and the orientation of every remaining docking port.
  3. S20.03Match each port's flange stack, rim shape, recess, and projected length from the node.
  4. S20.04Check occupied, capped, and available ports against the frozen scene manifest.
  5. S20.05Reproduce thermal blanket segments around the closely spaced port openings without intersecting the rings.
  6. S20.06Place photographed rails, external fittings, and cable interfaces in their correct positions.
  7. S20.07Inspect the Nauka junction from all reachable sides for alignment and proper structural contact.
  8. S20.08Match blanket and docking-ring materials with correct highlight width and shadowed detail.
  9. S20.09Compare six directional silhouettes; port symmetry must not erase real differences between individual interfaces.
  10. S20.10Orbit closely between the ports; reject missing back faces, intersecting collars, coarse curvature, and unstable seams.
21European Robotic Arm · ERARobotics · 10 checks · Unreviewed

Robot, end effectors, visible harnesses, and contact with its selected Nauka base points.

Reference anchors: R5. Matched photo boards and scores are pending.

  1. S21.01Match segment lengths, joint diameters, and the overall extended envelope to ESA specifications.
  2. S21.02Verify which end is attached and identify its exact Nauka base point in the chosen reference pose.
  3. S21.03Match the shoulder, elbow, and wrist angles in multiple views; perspective alone cannot determine a valid pose.
  4. S21.04Reconstruct both end effectors, cameras, and visible local mechanisms instead of using identical featureless caps.
  5. S21.05Inspect each joint connection for clean bearing gaps, correctly nested covers, and attached structural members.
  6. S21.06Match insulation wraps, seams, exposed joint metal, and reference markings without stretching textures around bends.
  7. S21.07Route visible harnesses through supported paths and inspect their clearance at each joint.
  8. S21.08Compare the arm against the Nauka silhouette, including clearance from arrays, radiator, and airlock.
  9. S21.09Inspect both ends and all joint undersides for disconnected meshes, pinched curves, and hidden intersections.
  10. S21.10Move the camera along the entire arm; small fittings and mechanical gaps must remain stable at every approach angle.
22Pressurized mating adapters and docking adaptersDocking · 10 checks · Unreviewed

PMA-1, PMA-2, PMA-3, IDA-2 and IDA-3, each with its own instance evidence and parent-port record.

Reference anchors: R1. Matched photo boards and scores are pending.

  1. S22.01Verify every adapter's identity, host port, direction, and relative rotation in the dated station manifest.
  2. S22.02Match each PMA's offset, taper, length, and end-ring proportions; do not replace it with a straight tube.
  3. S22.03Match each IDA's ring dimensions, rim profile, and attachment depth independently.
  4. S22.04Inspect mating planes for concentric contact, correct offsets, and no visible flange interpenetration.
  5. S22.05Reconstruct visible docking petals, guides, covers, and recesses at the level supported by close photographs.
  6. S22.06Match blanket segmentation, closure seams, and exposed structural edges on the individual adapter.
  7. S22.07Locate rails, cables, and external fittings without obstructing the docking opening.
  8. S22.08Compare black/white insulation and metal finishes under the same light as their parent modules.
  9. S22.09Inspect every adapter from the docking axis, side, underside, and its narrow rear junction.
  10. S22.10Fly past each rim and through any accessible passage; reject floating seals, missing interior walls, and flashing coplanar faces.
23Z1 truss and mounted equipmentTruss · 10 checks · Unreviewed

Z1 frame, control-moment-gyro housings, communications hardware, and fittings assigned to this host.

Reference anchors: R1 · R2. Matched photo boards and scores are pending.

  1. S23.01Match the frame's measured envelope and orientation above its actual node attachment.
  2. S23.02Reconstruct the observed beam cross-sections and diagonal pattern, including asymmetric frame bays.
  3. S23.03Locate each gyro housing and supporting mount from the selected configuration references.
  4. S23.04Inventory antenna and communications assemblies hosted here, recording dish, boom, and mount orientation separately.
  5. S23.05Match attachment feet and support paths between Z1 and the station without beams ending in empty space.
  6. S23.06Reproduce visible equipment boxes and covers by silhouette and mounting depth rather than random box scatter.
  7. S23.07Route observed cables and utility bundles through specific frame locations with supports.
  8. S23.08Match exposed structure, blanket-covered equipment, and antenna surfaces as different materials.
  9. S23.09Inspect all six directions and the interior of open bays for missing structure and unintended intersections.
  10. S23.10Fly beneath and through camera-accessible gaps; diagonal edges, cables, and small mounts must remain stable.
24S0 central trussTruss · 10 checks · Unreviewed

Central frame, laboratory support struts, rails, fixed utility hardware; mobile carriage is part 37.

Reference anchors: R2. Matched photo boards and scores are pending.

  1. S24.01Match the central truss dimensions, cross-section, and position relative to Destiny.
  2. S24.02Reconstruct the actual bay lengths and diagonal connections rather than repeating a generic lattice cell.
  3. S24.03Match laboratory support strut count, attachment points, and load-path geometry to reference views.
  4. S24.04Verify alignment and spacing of the mobile-transporter rail surfaces through the central section.
  5. S24.05Inventory fixed equipment boxes, covers, and utility interfaces on the correct faces.
  6. S24.06Match beam end plates, node connections, and visible fastener scale at astronaut approach distance.
  7. S24.07Route reference cable bundles and fluid lines without crossing open travel paths incorrectly.
  8. S24.08Match structural metal, white covers, and insulated equipment under common sun and Earth lighting.
  9. S24.09Inspect underside attachments and both lateral truss connections for gaps, overlapping beams, and unsupported rails.
  10. S24.10Traverse above, below, and alongside S0; thin diagonals and rail edges remain stable without obvious detail popping.
25P1 truss and port radiator bankTruss · 10 checks · Unreviewed

Port inboard structure, its thermal radiator assembly, fixed equipment, and communications fittings.

Reference anchors: R2 · R6. Matched photo boards and scores are pending.

  1. S25.01Match P1's length, cross-section, and connection orientation relative to S0 and P3.
  2. S25.02Reconstruct the correct bay pattern, beam profiles, and underside structure in matched views.
  3. S25.03Match each radiator panel's outline, segmentation, edge frame, and visible thickness independently.
  4. S25.04Verify radiator root hinges and the selected deployment angles against dated references.
  5. S25.05Inspect radiator supports and their connection to the truss; sheets cannot hover above the frame.
  6. S25.06Match thermal plumbing, cable routes, and equipment mounting arrangements along the port structure.
  7. S25.07Inventory antenna and other protruding fittings on this host and compare their local silhouettes.
  8. S25.08Match the radiator face/back contrast and frame material without making every surface uniformly white.
  9. S25.09Check the back of every radiator and the bank's clearance from adjacent equipment at grazing angles.
  10. S25.10Fly parallel to each radiator edge and beneath P1; no paper-thin silhouettes, flashing seams, or detached pipes.
26S1 truss and starboard radiator bankTruss · 10 checks · Unreviewed

Starboard inboard structure, thermal radiators, fixed equipment, and communications fittings.

Reference anchors: R2 · R6. Matched photo boards and scores are pending.

  1. S26.01Match S1 dimensions and its actual relation to S0 and S3 in the station layout.
  2. S26.02Reconstruct the starboard-specific bay structure; do not assume all port-side equipment is mirrored.
  3. S26.03Match every radiator's panel divisions, edge shape, thickness, and back structure to references.
  4. S26.04Verify hinge locations and radiator deployment angles for the frozen scene.
  5. S26.05Inspect the bank's supporting structure and each connection to the truss from below.
  6. S26.06Map visible pumps, boxes, plumbing, and cables to their correct starboard mounting positions.
  7. S26.07Check antenna, light, and protruding utility fittings against the host-specific inventory.
  8. S26.08Match radiator face, reverse surface, frame metal, and blanket material under consistent illumination.
  9. S26.09Compare the bank's full silhouette and clearance to nearby payloads from front, rear, top, and underside.
  10. S26.10Sweep along every radiator and through open visual gaps; reject intersecting sheets, floating mounts, and shimmering edges.
27P3/P4 truss and array jointsTruss · 10 checks · Unreviewed

Port inner photovoltaic structure, rotary joint, array roots, local thermal hardware; blankets are part 33.

Reference anchors: R2. Matched photo boards and scores are pending.

  1. S27.01Match P3/P4's structural envelope and its measured placement between P1 and P5.
  2. S27.02Reconstruct the correct transition between the truss sections and the photovoltaic structure.
  3. S27.03Match the solar alpha rotary joint's diameter, ring depth, housing detail, and rotation axis.
  4. S27.04Verify array-root and beta-joint locations against the identified port-side power channels.
  5. S27.05Match visible power-control equipment and its bracketed attachment to the structure.
  6. S27.06Inventory local radiator/thermal hardware and compare its deployed shape and mounting for the selected date.
  7. S27.07Route utility bundles around rotary interfaces with plausible attachment and clearance.
  8. S27.08Match joint metal, blanket covers, and exposed beams without uniform surface roughness.
  9. S27.09Inspect ring back faces, root joints, and the P5 connection for missing structure or mesh penetration.
  10. S27.10Compare a close orbit and a moving wide view; the joint stays concentric and the array roots remain visibly connected.
28S3/S4 truss and array jointsTruss · 10 checks · Unreviewed

Starboard inner photovoltaic structure, rotary joint, array roots, and local thermal hardware.

Reference anchors: R2. Matched photo boards and scores are pending.

  1. S28.01Match the section lengths, cross-sections, and placement between S1 and S5.
  2. S28.02Reconstruct the starboard-specific transition geometry and equipment distribution from references.
  3. S28.03Match rotary-joint ring diameter, thickness, covers, and its actual axis through the structure.
  4. S28.04Verify both array-root positions and beta-joint orientations using their channel identities.
  5. S28.05Match power equipment shapes and their support brackets at the correct structural bays.
  6. S28.06Inspect local thermal hardware, including its dated deployment and pipe connections.
  7. S28.07Match visible harness routing around the rotating interfaces without stretched or unsupported cables.
  8. S28.08Compare beam metal, joint housings, and insulated equipment under identical lighting to P3/P4.
  9. S28.09Inspect both sides of the rotary joint and the S5 interface for incorrect gaps and overlaps.
  10. S28.10Fly from the central truss past both array roots; the section must hold its silhouette without detail popping or joint wobble.
29P5 spacer trussTruss · 10 checks · Unreviewed

Port spacer frame, end interfaces, fixed cabling, and visible fittings.

Reference anchors: R2. Matched photo boards and scores are pending.

  1. S29.01Match spacer length, cross-section, and its gap-setting role between P4 and P6.
  2. S29.02Reproduce the exact diagonal pattern and open spaces in matched side and end views.
  3. S29.03Match beam profiles and end-node sizes instead of using identical round rods.
  4. S29.04Check both end interfaces for aligned mating plates and continuous attachment to neighboring structures.
  5. S29.05Place any visible fixed brackets and utility fittings in their documented bays.
  6. S29.06Route cables along their reference paths with support spacing visible at close range.
  7. S29.07Inspect internal beam crossings; joined members need plausible nodes, and unrelated members must not interpenetrate.
  8. S29.08Match bare metal and insulated surfaces without exaggerated rust, scratches, or procedural grime.
  9. S29.09Compare the negative space inside the frame from four oblique angles, not just its outer bounding box.
  10. S29.10Move a camera along the spacer and through open sightlines; reject disappearing diagonals and flickering cable detail.
30S5 spacer trussTruss · 10 checks · Unreviewed

Starboard spacer frame, end interfaces, cabling, and visible fittings.

Reference anchors: R2. Matched photo boards and scores are pending.

  1. S30.01Match spacer dimensions and placement between S4 and S6 using starboard references.
  2. S30.02Reconstruct the observed diagonal pattern and cross-bracing with correct open bay shapes.
  3. S30.03Match beam sections, corner nodes, and attachment-plate thickness at astronaut scale.
  4. S30.04Verify both end planes and their alignment with adjacent photovoltaic structures.
  5. S30.05Inventory local brackets and fittings rather than mirroring P5 without evidence.
  6. S30.06Match utility routing and cable support positions through the starboard spacer.
  7. S30.07Inspect beam junctions and internal crossings for impossible structural intersections.
  8. S30.08Compare metal highlights and local insulation with adjacent truss sections under common lighting.
  9. S30.09Check front, rear, underside, and end-on negative spaces against matched reference views.
  10. S30.10Fly along both edges and beneath the spacer; thin beams must remain continuous, stable, and correctly shaded.
31P6 outer photovoltaic trussTruss · 10 checks · Unreviewed

Outboard port structure, array roots, electrical equipment, and local radiator assemblies.

Reference anchors: R2 · R6. Matched photo boards and scores are pending.

  1. S31.01Match P6's full structural dimensions and placement beyond P5.
  2. S31.02Reconstruct its frame bays, equipment-bearing surfaces, and end geometry from multiple views.
  3. S31.03Verify both legacy array-root positions and orientations against their channel identities.
  4. S31.04Match local radiator outlines, panel divisions, and deployment state in the frozen configuration.
  5. S31.05Inspect radiator hinges, supports, and plumbing at their actual truss attachment points.
  6. S31.06Map electrical and battery equipment to the correct bays and reproduce distinctive covers.
  7. S31.07Match cable bundles, array-root fittings, and installed iROSA support interfaces without duplicate geometry.
  8. S31.08Compare radiator, insulation, beam metal, and electrical-cover materials under calibrated sunlight.
  9. S31.09Inspect the outboard end and underside for missing details hidden in station-wide photographs.
  10. S31.10Fly around the entire structure and between visible appendages; no panel intersections or thin-surface disappearance.
32S6 outer photovoltaic trussTruss · 10 checks · Unreviewed

Outboard starboard structure, array roots, power equipment, and local radiator assemblies.

Reference anchors: R2 · R7. Matched photo boards and scores are pending.

  1. S32.01Match the outer section's dimensions and its connection beyond S5.
  2. S32.02Reconstruct the actual frame bays and equipment distribution from starboard-specific photographs.
  3. S32.03Verify both legacy array-root positions and the associated iROSA mounting locations.
  4. S32.04Match local radiator shapes, segmentation, and deployed geometry for the selected date.
  5. S32.05Inspect radiator mounting hardware and utility connections from the reverse side.
  6. S32.06Match battery and electrical hardware shapes, cover seams, and supporting brackets.
  7. S32.07Route cables through documented supports without crossing solar blankets or open mechanical joints.
  8. S32.08Match radiator coatings, insulation, exposed frames, and dark equipment apertures as distinct materials.
  9. S32.09Compare the outboard silhouette and underside structure with matched close and wide references.
  10. S32.10Sweep around the section and its array roots; reject floating equipment, disconnected radiators, and shimmering framework.
33Eight legacy solar-array wingsPower · 10 checks · Unreviewed

All eight wings: separate evidence for 1A, 1B, 2A, 2B, 3A, 3B, 4A, and 4B; ten checks per instance.

Reference anchors: R2 · R7. Matched photo boards and scores are pending.

  1. S33.01Verify each wing's channel identity, host truss, side, and orientation in the station configuration.
  2. S33.02Match deployed length, total width, central mast, and blanket separation using dimensioned references.
  3. S33.03Reconstruct cell-grid proportions and blanket segmentation; a generic tiled checkerboard is a failure.
  4. S33.04Match the central mast, end bars, tension geometry, and root structure at close range.
  5. S33.05Inspect blanket thickness and edge silhouette without turning flexible blankets into thick rigid boards.
  6. S33.06Match front and rear surface appearance independently, including conductor and support patterns visible in sources.
  7. S33.07Verify the documented joint pose and clearance to adjacent wings, radiators, and iROSA overlays.
  8. S33.08Compare warm/cool color shifts and glints under matched light; reject neon orange paint or uniform metallic mirrors.
  9. S33.09Inspect grazing-angle motion for moire, aliasing, shadow acne, and unstable cell textures.
  10. S33.10Compare every complete wing from both sides and end-on; all eight must retain detail and connection continuity.
34Six iROSA wings and support kitsPower · 10 checks · Unreviewed

The six installed June 2023 wings on 2B, 4B, 3A, 4A, 1A, and 1B; separate evidence for every installation.

Reference anchors: R7. Matched photo boards and scores are pending.

  1. S34.01Verify the six correct channel locations and leave non-equipped channels free of invented installed wings.
  2. S34.02Match each roll-out wing's length, width, and relative footprint on the older array.
  3. S34.03Reconstruct the support-kit geometry and angle rather than placing a floating rectangle in front of a legacy wing.
  4. S34.04Match deployment booms, end structure, rolled-edge details, and root hardware from installation photographs.
  5. S34.05Reproduce the visible photovoltaic cell pattern at the correct scale and orientation.
  6. S34.06Match front/back appearance and edge thickness as separate observed surfaces.
  7. S34.07Trace power cables to their proper interfaces with attached connectors and supported routing.
  8. S34.08Check overlay clearance and shadow placement on the original array from several oblique angles.
  9. S34.09Inspect the full support kit from behind for disconnected beams, duplicate faces, and hidden intersections.
  10. S34.10Fly past each of the six installations; cell detail, thin booms, and contact shadows remain stable in motion.
35Canadarm2Robotics · 10 checks · Unreviewed

Arm, both end effectors, cameras, blankets, joints, and its visible attachment in the chosen pose.

Reference anchors: R8. Matched photo boards and scores are pending.

  1. S35.01Match overall reach, segment proportions, and joint sizes to CSA specifications and photographs.
  2. S35.02Verify both end identities and the actual attached grapple fixture in the frozen reference pose.
  3. S35.03Match shoulder, elbow, and wrist bends across multiple views; do not accept a pose inferred from one silhouette.
  4. S35.04Reconstruct each joint housing, bearing gap, and mechanical connection with clean, concentric geometry.
  5. S35.05Match end-effector depth, opening, mechanisms, cameras, and local fittings at close range.
  6. S35.06Reproduce blanket seams, lettering, and their orientation along the separate arm segments.
  7. S35.07Check rigid-segment texture continuity and harness routing without stretching across joints.
  8. S35.08Compare blanket whiteness, exposed metal, dark lenses, and wear under the same lighting as the station.
  9. S35.09Inspect every joint underside and the base contact for collisions with the station or detached arm segments.
  10. S35.10Orbit from root to tip and compare a full-station view; the arm retains correct scale and clear silhouette throughout.
36DextreRobotics · 10 checks · Unreviewed

Torso, both manipulator arms, tools, cameras, and its dated mount/storage pose.

Reference anchors: R9. Matched photo boards and scores are pending.

  1. S36.01Match torso size, arm proportions, and overall envelope to CSA reference dimensions.
  2. S36.02Verify the selected mounting location and the mechanical connection to its carrier or arm.
  3. S36.03Match both arm poses independently, including shoulder, elbow, and wrist orientation.
  4. S36.04Reconstruct joint housings and the gap geometry that distinguishes articulated mechanisms from joined cylinders.
  5. S36.05Match tool interfaces and end mechanisms rather than terminating each arm in a featureless block.
  6. S36.06Place cameras, work lights, tool holders, and visible auxiliary equipment from a dated inventory.
  7. S36.07Match insulation seams, exposed structure, and recognizable markings at close approach distance.
  8. S36.08Inspect harnesses and mechanical attachments behind the torso and between the arms.
  9. S36.09Compare the front, rear, underside, and both sides for accidental self-intersection and missing equipment.
  10. S36.10Fly around the full robot against station and Earth backgrounds; fine tools must remain attached and temporally stable.
37Mobile Base System and Mobile TransporterRobotics · 10 checks · Unreviewed

Platform, carriage, rail contacts, grapple fixtures, and fixed hardware; carried robots are scored separately.

Reference anchors: R10. Matched photo boards and scores are pending.

  1. S37.01Match platform dimensions, height, and outline against CSA specifications.
  2. S37.02Verify its selected rail worksite and orientation along the integrated truss.
  3. S37.03Reconstruct the carriage's rail contacts and support geometry without a platform floating above the track.
  4. S37.04Match the four grapple fixtures, their orientations, mounting plates, and spatial arrangement.
  5. S37.05Inspect the payload support interfaces and visible hold-down hardware from close reference views.
  6. S37.06Match platform deck, framework, cable routing, and equipment boxes on both upper and lower faces.
  7. S37.07Check clearances to adjacent truss equipment and the carried robot in the frozen pose.
  8. S37.08Compare exposed structural metal, white covers, and dark mechanical regions under consistent light.
  9. S37.09Inspect the underside and wheel/rail region for duplicated tracks, missing contacts, and intersecting geometry.
  10. S37.10Fly along the rail and around the platform; grapple detail and carriage supports remain readable and stable.
38EXPRESS Logistics Carriers · ELC-1 through ELC-4External science · 10 checks · Unreviewed

Every carrier, its individual payload manifest and mounting structure; all four assessed separately.

Reference anchors: R1. Matched photo boards and scores are pending.

  1. S38.01Verify each carrier's identity, host truss, attachment face, and rotational orientation.
  2. S38.02Match each platform's outline, thickness, frame, and attachment geometry to photographs.
  3. S38.03Map occupied and empty payload positions independently for the frozen date.
  4. S38.04Reconstruct each carried unit's visible shape, proportions, cover seams, and orientation rather than generic box clutter.
  5. S38.05Match payload feet, restraint fixtures, and carrier interfaces so every unit has a visible supported attachment.
  6. S38.06Trace exposed utility connections and cables to the correct payload and carrier interfaces.
  7. S38.07Inspect every platform underside, edge, and support bracket in oblique views.
  8. S38.08Match blankets, radiator-like faces, metal frames, and dark apertures by payload instead of applying one shared finish.
  9. S38.09Compare platform silhouettes and clearance from nearby arrays, robots, and station structure.
  10. S38.10Fly around all four carriers; no hidden duplicate payloads, floating boxes, overlapping mounts, or unstable thin hardware.
39External Stowage Platforms · ESP-1 through ESP-3External science · 10 checks · Unreviewed

All three platforms, their separate dated spare-equipment inventories, restraints, and supports.

Reference anchors: R1. Matched photo boards and scores are pending.

  1. S39.01Verify each platform's identity, location, mounting orientation, and attachment to its actual host.
  2. S39.02Match the individual platform outline, thickness, frame pattern, and support geometry.
  3. S39.03Inventory the stored units and empty attachment sites for the selected date.
  4. S39.04Match each spare unit's distinctive shape and dimensions; repeated generic crates fail.
  5. S39.05Reconstruct restraint and attachment hardware with visible contact between the stored unit and platform.
  6. S39.06Match protective covers, wrapped equipment, handles, and visible identification markings to photographs.
  7. S39.07Inspect rear surfaces, platform undersides, and narrow gaps between stored units.
  8. S39.08Compare bare structure, fabric covers, reflective insulation, and painted parts as separate materials.
  9. S39.09Check the complete stowage silhouette and clearance from nearby astronaut routes and station appendages.
  10. S39.10Perform a close flyby of every platform; reject unsupported spares, penetrating straps, missing back faces, and unstable edges.
40Alpha Magnetic Spectrometer · AMS-02External science · 10 checks · Unreviewed

Instrument envelope, external thermal hardware, support structure, and equipment visible in June 2023.

Reference anchors: R11. Matched photo boards and scores are pending.

  1. S40.01Match the complete instrument's measured envelope and distinctive multi-part silhouette.
  2. S40.02Verify its mounting position and orientation on the starboard truss.
  3. S40.03Reconstruct the primary housing, top geometry, and exposed structural framework from multiple photographs.
  4. S40.04Match thermal panels and external radiator/support geometry, including visible reverse surfaces.
  5. S40.05Inventory external boxes, blankets, and changes present by the chosen date; launch configuration alone is insufficient.
  6. S40.06Match cable and cooling-line routes, connectors, and their supports without arbitrary dangling detail.
  7. S40.07Inspect the attachment structure and all load-bearing contacts with the station from underneath.
  8. S40.08Compare foil, white thermal surfaces, exposed metal, and dark apertures under calibrated illumination.
  9. S40.09Check its silhouette and clearances against neighboring payloads, truss members, and solar-array hardware.
  10. S40.10Orbit the complete instrument at close and medium range; layered covers and support members remain clean and stable.

No matching station parts. Try a different word or group.

Six additional acceptance gates

These 60 checks are separate from the 400 station checks. A beautiful station alone does not establish that the game meets your request.

A · Preserve the astronaut; add convincing animation — 10 checks

Reference anchors: R15

  1. A.01Preserve the approved suit's silhouette, fabric, patches, visor, gloves, boots, and color. Compare the same ten baseline views before and after rigging; any visible loss requires correction.
  2. A.02Add an articulated skeleton with anatomically placed shoulders, elbows, wrists, hips, knees, and ankles. Inspect shaded and wireframe deformation at neutral and extreme allowed poses.
  3. A.03Make arms and legs visibly respond during movement, with restrained motion appropriate to a bulky pressurized suit; reject a rigid statue sliding through the scene.
  4. A.04Use gentle, asymmetric arm and leg drift at rest. Limbs settle smoothly after input stops, without synchronized pendulum loops or sudden resets.
  5. A.05Use a grounded stepping cycle only when feet have a supported contact surface; free flight uses floating poses, with no walking through empty space.
  6. A.06Keep fabric volume across bent elbows, shoulders, knees, and hips. Reject collapsing joints, torn seams, stretched patches, and candy-wrapper twisting.
  7. A.07Keep gloves attached at the wrist and boots attached at the ankle through the full pose range; no hand, sleeve, or knee passes through another suit surface.
  8. A.08Keep the backpack and helmet rigidly mounted while the body animates; neither may squash with skin weights or penetrate the torso.
  9. A.09Attach the tether to its visible suit fitting and update it through every pose. Inspect the front, rear, underside, and full camera orbit while moving.
  10. A.10Record forward, reverse, sideways, vertical, braking, idle, and turning sequences. Preserve appearance in motion and confirm the animated envelope stays clear of station geometry.
C · Camera, movement, view buttons, and tether — 10 checks
  1. C.01Right-button drag orbits the camera independently around the astronaut through a full 360 degrees, including an unobstructed front/visor view and a visible tether attachment.
  2. C.02Right-drag changes neither body heading nor thrust direction. Suppress the browser context menu only on the game canvas; releasing the button leaves the chosen view in place.
  3. C.03Left-button drag turns the astronaut's body and its forward movement heading. WASD supplies translation; the camera follows continuously without sudden snapping.
  4. C.04Keep vertical movement and braking available with clearly displayed bindings. Translation, turning, and camera orbit must work in combination without stuck input.
  5. C.05Provide an explicit recenter control and bounded camera distance. Vertical orbit must not flip the image or pass through the astronaut's helmet and body.
  6. C.06Preserve the existing named view buttons and make their effects clear. Interior, EVA, station overview, and suit inspection must load a readable, unclipped composition.
  7. C.07Allow inspection around both ends, all sides, and beneath the entire station. Camera collision should slide or shorten smoothly without passing through structure or hiding the astronaut unnecessarily.
  8. C.08Keep the tether attached at both visible ends, routed around obstacles, and within its allowed length. Slack must not sag as if under Earth-surface gravity or pass through the suit.
  9. C.09Support a usable route around the station through tether reattachment/reeling and free movement as applicable. Test constrained corners, full extension, detachment, return, and recovery from obstruction.
  10. C.10Test button release outside the canvas, lost focus, simultaneous buttons, view switching, and reload. Controls must recover cleanly; use isolated Playwright self-play and captured visual evidence.
E · Earth detail, clouds, atmosphere, and lighting — 10 checks

Reference anchors: R12

  1. E.01Choose imagery by ground detail visible from the actual orbital altitude. A larger file or an 8K label does not pass: compare coastlines, terrain, and cloud edges at native 1080p and 1440p captures.
  2. E.02Use sufficient source detail and distance-appropriate tiles so close Earth views have no obvious square pixels or smeared continents. Upscaled low-resolution imagery alone fails.
  3. E.03Inspect tile boundaries, longitude wrap, poles, coastlines, and ocean/land transitions for seams, mismatched color, duplicate clouds, and blurred joins.
  4. E.04Build clouds as a separate altitude layer with convincing thickness, soft edges, and varied structure. Reject clouds painted onto the land or an opaque white shell.
  5. E.05Animate clouds slowly relative to the ground with regional variation and continuous motion; rotating one rigid cloud globe is insufficient for the requested improvement.
  6. E.06Show cloud motion in a time-separated comparison with ground landmarks tracked. At normal speed it is subtle; time acceleration must reveal continuous drift without a looping jump.
  7. E.07Match the thin atmospheric limb and horizon falloff to orbital photographs, using an actual altitude-to-radius relationship rather than an oversized neon halo.
  8. E.08Use one coherent Sun direction for Earth, clouds, station, and astronaut. Cloud shadows, land shading, and the day/night boundary must agree.
  9. E.09Match ocean darkness, land saturation, cloud brightness, and glint to exposure-matched references. Preserve cloud detail without crushed land or blown-out white surfaces.
  10. E.10Capture horizon, straight-down, sunset, and night-side motion. Reject cloud/ground intersections, texture swimming, unstable shadows, visible detail transitions, and excessive stars under daylight exposure.
O · Real orbital motion and a stable playable station — 10 checks

Reference anchors: R13

  1. O.01Use metres consistently and record Earth radius, orbital altitude, inclination, and simulation time in the scene specification. The station's scale must remain 1:1.
  2. O.02Move the station's reference frame along an actual orbital trajectory around Earth; spinning the Earth texture under a stationary station alone fails.
  3. O.03Calculate orbital period from the chosen radius and gravitational parameter. Verify a complete simulated orbit against the calculation and the reference altitude, rather than hardcoding an arbitrary fast loop.
  4. O.04Advance Earth's rotation separately from the station orbit and cloud drift. Track a named ground landmark to demonstrate these motions are independent.
  5. O.05Keep the astronaut, station, active camera, and tether in a numerically stable local frame that follows the orbit together; no drift away from the station when controls are idle.
  6. O.06Use a documented station attitude relative to the orbit so Earth-facing views remain physically coherent. Compare the horizon direction from the Cupola and exterior views.
  7. O.07Keep Sun direction consistent through orbital motion; station shadows, illuminated array faces, Earth lighting, and eclipse timing must agree.
  8. O.08Offer clearly labeled real-time and accelerated orbital viewing. Acceleration changes simulation time consistently and must not secretly change astronaut movement speed.
  9. O.09Switch views, orbit the astronaut camera, and attach/reel the tether during accelerated and normal orbit. Reject jumps, precision jitter, length changes, or broken relative positions.
  10. O.10Record continuous normal-speed and accelerated sequences plus positions over one orbit. Require visible ground-track progression and correct shadow changes; a static screenshot cannot establish this gate.
I · Interior fidelity and continuity — 10 checks

Reference anchors: R3 · R4 · R14

  1. I.01Replace any invented straight corridor layout in explorable areas with the actual connected-module topology and measured hatch-center relationships.
  2. I.02Match interior cross-sections, passage widths, rack depths, and hatch sizes to the specific module; one corridor template repeated everywhere fails.
  3. I.03Build reference boards for every player-accessible interior wall and end, including the ceiling and areas behind the starting camera.
  4. I.04Match racks, storage bags, restraints, ventilation grilles, and labels by location and scale; equipment density must follow references rather than random clutter.
  5. I.05Attach cables, bags, panels, and equipment to visible supports. Inspect their back sides and every accessible gap for floating objects and intersections.
  6. I.06Use distinct textiles, coated panels, metal fittings, displays, and glazing. Match local illumination without flattening the cabin into uniform white or colored neon.
  7. I.07Preserve clean sightlines and astronaut clearance at actual scale; adjust navigation and animation instead of secretly widening the station.
  8. I.08Keep interior windows and exterior openings physically consistent, including frame thickness and the correct view of Earth or station equipment.
  9. I.09Allow the existing view-based inside/outside transition; door mechanics are optional. Arrival pose, camera clearance, controls, and tether state must remain coherent.
  10. I.10Record a continuous tour of every explorable interior area, looking at all walls and moving through each junction. No walk-through walls, clipped suit, missing ceiling, or exposed exterior shell.
G · Whole-scene acceptance and non-regression — 10 checks
  1. G.01Compare full-station forward, aft, port, starboard, zenith, and nadir renders against the frozen configuration, then repeat matched oblique flyaround views. Wrong topology or a missing assembly blocks completion.
  2. G.02Maintain a host-and-child inventory for every visible component, including antennas, visiting vehicles, science payloads, brackets, and transferred hardware. Every item must belong to a rubric and have reference evidence.
  3. G.03Run named collision approaches against every major assembly and its protrusions, in both travel directions where reachable. Test the astronaut, animated limbs, camera, and tether separately.
  4. G.04Use native-resolution stills and moving captures at 1080p and 1440p. Inspect a screen grid for silhouette errors, blur, clipped details, floating objects, and excessive post-processing.
  5. G.05Keep normals, tangents, triangulation, glass ordering, shadows, texture filtering, and detail transitions stable. Zero reproducible visible cracks, exposed intersections, flicker, or missing back faces are allowed on reachable views.
  6. G.06Use a recorded reference device/browser/settings profile. Proposed performance gate: median at least 60 FPS, p95 frame time at most 22 ms, and p99 at most 33.3 ms on each warmed 120-second 1080p traversal. Report 1440p separately.
  7. G.07Measure complete traversals around arrays, radiators, robotics, interiors, Earth, and the animated suit. No traversal may be represented by an empty-scene or stationary-camera benchmark.
  8. G.08Record cold-load time, asset bytes, memory, loading failures, and console errors. Required assets must load, no fatal errors are allowed, and warmed movement must have no asset-loading freeze over 100 ms.
  9. G.09Compare every candidate with the accepted baseline at identical cameras, lighting, settings, and motion. Improvements cannot conceal regression in the approved suit, input, clarity, collision coverage, or performance.
  10. G.10Accept only the exact built version represented by the complete evidence ledger. Publish or hand over a final-quality claim only after all required checks pass; list remaining defects explicitly on any requested interim preview.

One additional branding check

B.01 — Authentic OpenAI blossom on both astronaut and ISS: preserve the official mark proportions and orientation; place it on a physically attached suit patch and a station plaque; preserve existing mission badges; keep both legible at inspection distance with matching surface material, clean edges, no mirroring, no floating, clipping or flicker in all reachable angles and animated poses. Record separate suit and station evidence; both must pass. These requested marks are intentional artistic additions, not historical ISS or NASA branding.

Separate child evidence for the suit patch and station plaque; the lower result controls B.01.

Evidence travels with the result

Each check begins as Unreviewed. After implementation, its ledger row links the exact reference, build, camera, screenshot or motion clip, discrepancy, and retest. No pre-filled passing scores.

Check ID · Child instance · Build ID · Source image IDs and dates
Camera + lighting · Screenshot / clip · Measurement · Score
Status · Defect · Retest evidence

Download the 461-row evidence ledger

Reference starting points

Official agency references anchor the approved breakdown. They are not yet a completed set of matched photo boards. Capture dates, hardware changes, pose, and source precision must be reconciled before an individual comparison passes.

R1NASA · Station assembly elements

Assembly names and dated installation history. Individual dated close-up photographs still need to be selected for each check.

R2NASA · Integrated truss structure

Truss organization and reference entry point; appendage pose requires dated imagery.

R3JAXA · Kibo

Japanese laboratory reference entry point. Archive diagrams must be checked for equipment no longer present.

R4ESA · European contributions

Useful historical module reference; future-tense launch descriptions do not establish the June 2023 configuration.

R5ESA · European Robotic Arm

Arm description and specifications; use on-orbit photographs to select its pose.

R6ESA · ISS radiators, Luca Parmitano photograph

Close radiator construction reference, photographed in 2019. Confirm pose separately for the scene date.

R7NASA · Sixth iROSA installation, 15 June 2023

Dated anchor for the proposed six-iROSA station configuration.

R8CSA · Canadarm2

Official specifications and image/video entry points.

R9CSA · Dextre

Official robot reference and image/video entry points; exclude later upgrades from the chosen scene.

R10CSA · Mobile Base System

Platform, rail-car relationship, and dimensional reference.

R11NASA · Alpha Magnetic Spectrometer

Instrument reference; external service changes need dated photographs.

R12NASA · Blue Marble base-map source tiles

500 m source tiles are a candidate for sharper land imagery; this is historical imagery, not live weather or proof of final visual quality.

R13NASA · Station facts and figures

Orbital context; the simulation must use explicitly recorded values rather than imply live ISS tracking.

R14NASA · Destiny robotics workstation

Interior layout and equipment reference for the photographed area, not an entire-interior reconstruction.

R15Smithsonian · Armstrong spacesuit scan

Approved historical suit appearance is retained by request. The avatar is an explicit artistic exception to a period-accurate ISS crew suit.

R16NASA · ISS(C) high-resolution 3D model candidate

Candidate to inspect after review. A NASA model or high-resolution label does not automatically pass the geometry, date, texture, or close-up checks.