Sendsei AI game review
Access the Inaccessible · Live CV Review
Real video · Gabo CV cv-0.2.0 · evidence-grounded analysis
Evidence quality medium
Your attempt
0:000:09
Analyzed window0:00–0:098.0s descent excluded
AI analysis engine
Analysis
Grounded in all 20 measurements across 5 phases. Symbols classify evidence—not difficulty or send probability.
Review · P1
Establish
Establish phase. Left hand confirmed on LONG-H1 (confidence 0.88). Right hand visibility low (0.076) but contacts from later phases show it reaches LONG-H3. Both feet low visibility. Hip pause 1300 ms with drift 0.054.
Active contacts
Left Hand · LONG-H1 · 88%
CV flags
No unusual signal
Evidence noteright hand visibility below 0.5 for 41 frames
Evidence filmstrip
Eight frames behind the engine call.
Select a frame to seek the analyzed video. These stills are selected by the deterministic CV pipeline.
Confirmed color-locked route
8 holds drive this analysis.
Observed contact sequence: LONG-H1 → LONG-H3 → LONG-H2 → LONG-H5. No optimal beta is claimed without a confirmed reference attempt.
These are image-plane estimates, not an optimal beta. Wall angle, depth, grip quality, force, and fatigue are not modeled.Current phase · Complete measurements
P1 exposes every CV metric.
Durations, pauses, drift, sway, and regrabs come from the deterministic pipeline. They are image-plane proxies—not force or definitive cause.
Wall-clock time between the contact-state changes bounding this phase.
Longest interval with projected hip speed below threshold; 2D image-plane measure.
Net 2D displacement of the hip midpoint over the phase; not a 3D or center-of-mass measurement.
Total 2D hip path length divided by net displacement over the phase; 2D image-plane measure, only emitted when net displacement >= 0.01.
Priority momentP2
The climber established left hand on LONG-H1 (start), then reached right hand to LONG-H3, followed by left hand to LONG-H3 (via candidate LONG-H2), and finally right hand to finish hold LONG-H5 with a regrab. The attempt ended at the finish hold. The divergence occurred in P2, where the left hand reached to LONG-H1 but the climber exhibited a long dwell (3334 ms) and high hip sway (ratio 3.12) while both feet had low visibility, suggesting difficulty stabilizing after the first move.
Observed
In phase P2 (1333-4667 ms), left hand holds LONG-H1, right hand reaches LONG-H3, but the phase duration (3334 ms) is over 2x median with a hip pause of 2166 ms, and the hip path/displacement ratio (3.12) exceeds threshold 3.0, while left and right foot visibility are below 0.5 for extended periods.
Inferred
The climber likely struggled to stabilize after the initial left-hand hold, possibly due to poor foot placement or body tension, leading to prolonged adjustment and sway. However, low foot visibility limits confidence.
Other plausible explanations
- Climber may have been reading the next hold or resting rather than struggling.
- Low foot visibility could indicate occluded markers, not necessarily poor foot contact.
Known limitations
What this review cannot tell you.
- All measurements are 2D image-plane observations from one fixed camera; depth, force, and definitive causality are not observable.
- Standalone local analysis labels the required primary contract attempt as failed; the recording outcome was not independently verified.
- The route polygons were approximated from the supplied recording for this local run and require climber confirmation before production use.
- failed attempt: Analysis truncated at top-out (t=9500 ms): right hand held finish hold LONG-H5 for 1000 ms; the descent is not analyzed.
- Foot visibility was low in critical phases (P1, P2, P3), limiting assessment of foot placement and stability.
- Left hand on LONG-H2 during P4 had only 11 overlapping frames (<15), making that contact a candidate rather than confirmed.