Skip to content
Skills Library
Programming · Testing

camera-debugging-qa

/camera-debugging-qa

Use to measure camera jitter, follow errors, bounds, occlusion and teleport regressions from traces; not to build a camera rig.

Licence
MIT
Last repository change
30.10.2026

Assessment and security

100/ 100Class: good
Format25 / 25
Frontmatter with name and description, valid name, description up to 1,024 characters, name matching the folder.
Discoverability25 / 25
Description explains when to use the skill, has a reasonable length and uses third person.
Content35 / 35
At least 150 words and at most 500 lines, headings, step lists, examples or code.
Hygiene15 / 15
No placeholders such as TODO, no Windows paths, with a licence.
Riskno signals detected in the scan
Files in this package
12
Scripts
2
HTML comments
0
URLs in SKILL.md
0
Deprecation

No signals detected. Repository changed on 30.10.2026, without references to retired models or APIs.

Licence

MIT, open: use and modify while retaining author attribution.

What to check before installing
  • 2 scripts in this package. Review them before installing, as Claude may execute them.

Scripts: scripts/camera_trace_qa.py, tests/test_camera_trace_qa.py

What it does and when to use it

Use to measure camera jitter, follow errors, bounds, occlusion and teleport regressions from traces; not to build a camera rig.

Original author description

Use to measure camera jitter, follow errors, bounds, occlusion and teleport regressions from traces; not to build a camera rig.

How it works

  1. Procedure
  2. Built-in tools
  3. Measurement limitations
  4. Verification
Original SKILL.md headings · 34 lines, 399 words
SKILL.md preview
---
name: camera-debugging-qa
description: Use to measure camera jitter, follow errors, bounds, occlusion and teleport regressions from traces; not to build a camera rig.
---
# Camera Debugging & QA

Diagnose camera behavior with falsifiable metrics and scene reproduction. Do not assert visual comfort or GPU performance based only on numeric traces.

## Procedure

1. State the camera's intended behavior and operating conditions (platformer look-ahead, RTS pan, target group, scripted shot). Select measurements appropriate to that design; a camera deliberately framing *ahead* of the player will not have zero follow error.
2. Collect telemetry from **the existing render loop**. Write one record per rendered frame: timestamp `t` in seconds, world `target` and visual `camera` positions, optional `teleport`, optional `occluded`. Never store user-identifying data.
3. For 2D bounds, record the effective logical viewport, zoom, and rectangular level bounds. Match units: world positions and viewport divided by zoom. Do not conflate CSS pixels with backing pixels.
4. Save as JSONL, JSON array / `{ "records": [...] }`, or CSV with flat fields; see `references/trace-schema.md`.
5. Run `python scripts/camera_trace_qa.py TRACE.jsonl --max-speed 500 --max-velocity-jump 60 --output report.json` (thresholds illustrative, must be adapted per game). Exit 0=pass, 2=threshold breach, 3=bad input.
6. Review failure frames and reproduce them visually. Rerun the same seed, scene, machine class and input trace after change. Include 30/60/144 Hz or frame-time variation, direction reversal, pause/resume, resize, teleports and obstructions.
7. Report summary, pass/fail thresholds, any intentional violations and any GPU/browser/mobile tests not performed.

## Built-in tools

- `scripts/camera_trace_qa.py` — standard-library Python 3, JSONL/JSON/CSV, robust input validation, p50/p95 following error, speed, velocity jump, occlusion and bounds.
- `tests/test_camera_trace_qa.py` — automated edge-case tests.
- `examples/trace.jsonl` — known-good demonstration trace.
- `references/trace-schema.md` — schema, metric semantics and constraints.
- `assets/camera-qa-report.md` — reusable report.

## Measurement limitations

- Velocity jump is `|v[n]-v[n-1]|`, **not** acceleration per second squared and **not** jerk.
- A camera is not meant to sit on its target for all game styles; following error is diagnostic, not universally a failure.
- Samples recorded directly after teleports are excluded from follow/speed checks for a configurable grace interval; other failures are still tracked.
- Occlusion flag requires a **real visibility or collision query** supplied by the game; this tool cannot infer it from camera/target positions alone.
- Unit tests alone cannot prove screen composition, image stability, comfort, input latency or CPU/GPU performance.

## Verification

Run `python -m unittest discover -s tests -v` from the skill root; invoke CLI on the example trace; inspect JSON report. Validate that missing fields, NaN, non-monotonic timestamps and impossible bounds fail closed.
Files in this package (12)
  • CHANGELOG.md 1 KB
  • LICENSE 1 KB
  • PROVENANCE.md 1 KB
  • README.md 1 KB
  • SKILL.md 3 KB
  • assets/camera-qa-report.md 1 KB
  • attribution/upstream/LICENSE 11 KB
  • attribution/upstream/NOTICE 1 KB
  • examples/trace.jsonl 1 KB
  • references/trace-schema.md 2 KB
  • scripts/camera_trace_qa.py 9 KB
  • tests/test_camera_trace_qa.py 3 KB

Versions

The database has no other skills named camera-debugging-qa.

Database snapshot: 13.11.2026 · package version debbdba22f42 (SHA-256 digest of the entire package)

Downloads in the last 30 days: 0 · ranking

Report a problem with this skill: harmful content, incorrect descriptions or an author’s removal request.

Similar skills

  • 100camera-system
    Use when implementing cameras — smooth follow, screen shake, camera zones, and transitions for 2D and 3D
  • 100camera-system
    Use when implementing cameras — smooth follow, screen shake, camera zones, and transitions for 2D and 3D
  • 100browser-camera-2d
    Use for Canvas 2D or pixel-art camera follow, deadzones, look-ahead, zoom, bounds and teleport handling; not 3D camera rigs.
  • 100camera-systems
    Build game cameras that feel good — 2D follow with a deadzone, look-ahead, smoothing, and level-bounds clamping; 3D third-person orbit with collision and first-person look; plus multi-target framing and a shake hook. Engine-neutral techniques that pair with the engine's camera…
  • 100camera-systems
    Build game cameras that feel good — 2D follow with a deadzone, look-ahead, smoothing, and level-bounds clamping; 3D third-person orbit with collision and first-person look; plus multi-target framing and a shake hook. Engine-neutral techniques that pair with the engine's camera…
  • 90esp32-camera-xclk-timing-validation
    Use when designing, reviewing, debugging or validating camera xclk timing validation for ESP32 firmware; apply target-specific constraints, failure analysis and measurable verification before release.

Feedback and reports

Your message goes to Bartosz Sokoliński. An email address is needed only for a reply. See the privacy policy.