Quality assurance for AI-blurred videos — detect mask bleeding, temporal flicker, missed detections, edge halos, and motion-tracking failures. Use when user mentions blur quality, QA review, mask artifacts, flickering blur, missed faces, plate tracking failure, blur edge halo, or validating BGBlur output before delivery.
Works with
AI-first code editor with Composer
Before installing skills in Cursor, ensure your development environment meets these requirements:
node --versionvideo-blur-qaExecute the skills CLI command in your project's root directory to begin installation:
Fetches video-blur-qa from whyashthakker/bgblur-video-skills and configures it for Cursor.
The CLI shows a list of agents. Use arrow keys and space to select Cursor:
Confirm successful installation by checking the skill directory location:
Restart Cursor to activate video-blur-qa. Access via /video-blur-qa in your agent's command palette.
We perform automated surface-level scans (Gen AI Scanner, Socket, Snyk) during installation. These checks detect common vulnerabilities but do not guarantee complete security. Always review skill source code and verify the publisher's reputation before production use.
Skills execute code in your environment. Always review source, verify the publisher, and test in isolation before production.
Submit your Claude Code skill and start earning
Automate repetitive workflows and reduce manual effort
Example
Generate reports, summarize documents, draft communications
Save 3-5 hours per week on routine tasks
Learn new skills, understand complex topics, get expert guidance
Example
Explain concepts, provide examples, suggest learning resources
Accelerate learning and skill development by 2x
Enhance output quality through reviews, suggestions, and refinements
Example
Review drafts, suggest improvements, catch errors
Improve work quality by 30-40% with less effort
0
total installs
0
this week
0
upvotes
Run in your terminal
0
installs
0
this week
—
stars
| name | video-blur-qa |
| description | Quality assurance for AI-blurred videos — detect mask bleeding, temporal flicker, missed detections, edge halos, and motion-tracking failures. Use when user mentions blur quality, QA review, mask artifacts, flickering blur, missed faces, plate tracking failure, blur edge halo, or validating BGBlur output before delivery. |
| argument-hint | blurred video path, blur type (face/background/plate), or QA severity level |
| allowed-tools | Read, Write, Shell |
Systematic quality assurance for videos processed with BGBlur AI blur. Catch the failures users notice: flickering masks, missed faces, plate slips, and background bleed into subjects.
Top 5 blur defects:
Extract frames at high-risk timestamps:
python3 scripts/sample_frames.py "blurred_output.mp4" --output ./qa_frames/
Auto-samples: start, end, every 5s, and scene-change intervals.
Manual extraction at suspect timestamps:
ffmpeg -i blurred_output.mp4 -ss 00:01:23 -vframes 1 qa_frame_0123.jpg
Face Blur / Anonymization:
License Plate Blur:
Background Blur:
Blur Anything (prompt-based):
Review these high-risk segments at 2x playback:
| Segment Type | What to Check |
|---|---|
| Fast pan | Background blur edge stability |
| Subject turns head | Face mask follows rotation |
| Vehicle passes | Plate tracked through motion blur |
| Scene cut | New detections within 2 frames |
| Zoom in/out | Mask scale matches subject |
| Low light / noise | Detection doesn't drop out |
Compare original vs blurred for delivery QA:
ffmpeg -i original.mp4 -i blurred_output.mp4 \
-filter_complex "[0:v][1:v]hstack=inputs=2" \
-c:v libx264 -crf 18 comparison.mp4
python3 scripts/blur_qa_report.py "blurred_output.mp4"
Reports: resolution consistency, frame count, duration match, black frames, frozen segments.
| Severity | Definition | Action |
|---|---|---|
| P0 — Blocker | Unblurred PII visible (face, plate, screen) | Re-process; do not ship |
| P1 — Major | Tracking slip > 5 frames or identity reconstructable | Re-process affected segment |
| P2 — Minor | Edge halo, 1-2 frame flicker | Accept or touch up if client-facing |
| P3 — Cosmetic | Slight blur intensity inconsistency | Accept |
| Defect | Likely Cause | Fix |
|---|---|---|
| Face missed at cut | Scene change | Re-upload; trim at cut point and process separately |
| Plate slip | Fast motion / low res | Upscale source or trim to slower segment |
| Background eats hair | Similar color to bg | Reduce blur strength; improve subject/background contrast in source |
| Flickering mask | VFR source footage | Re-prep with ffmpeg-video-prep (force 30fps CFR) |
| Object not found | Vague prompt | Use specific prompt: "white Tesla license plate" not "plate" |
## Blur QA Report
### Asset
- File: [blurred_output.mp4]
- Blur type: [face / plate / background / object]
- Duration reviewed: [full / segments]
### Findings
| Timestamp | Severity | Issue | Notes |
|-----------|----------|-------|-------|
| 00:01:23 | P0 | Unblurred face | Bystander at frame edge |
| 00:02:45 | P2 | Edge halo | Subject hair, 3 frames |
### Verdict
- [ ] PASS — ready for delivery
- [ ] FAIL — re-process required
### Re-process Notes
[Specific segments, BGBlur mode changes, or prep fixes needed]
Re-process failed segments at BGBlur Upload. For systematic failures on long footage, consider Enterprise batch pipelines.
Prerequisites
Time Estimate
15-45 minutes depending on use case complexity
Steps
Common Pitfalls
✓ Do
✗ Don't
💡 Pro Tips
✓ Use when
Use when skill capabilities match your task, clear ROI on time saved, and you can validate outputs. Best for repetitive tasks, learning, and quality improvement.
✗ Avoid when
Avoid when task requires deep expertise you can't validate, involves sensitive decisions, or when learning process is more valuable than speed of completion.
whyashthakker/agent-skills-marketing
affaan-m/everything-claude-code
supercent-io/skills-template
heygen-com/skills
remotion-dev/skills
davila7/claude-code-templates
video-blur-qa reduced setup friction for our internal harness; good balance of opinion and flexibility.
video-blur-qa reduced setup friction for our internal harness; good balance of opinion and flexibility.
I recommend video-blur-qa for anyone iterating fast on agent tooling; clear intent and a small, reviewable surface area.
I recommend video-blur-qa for anyone iterating fast on agent tooling; clear intent and a small, reviewable surface area.
Useful defaults in video-blur-qa — fewer surprises than typical one-off scripts, and it plays nicely with `npx skills` flows.
Useful defaults in video-blur-qa — fewer surprises than typical one-off scripts, and it plays nicely with `npx skills` flows.
video-blur-qa reduced setup friction for our internal harness; good balance of opinion and flexibility.
We added video-blur-qa from the explainx registry; install was straightforward and the SKILL.md answered most questions upfront.
We added video-blur-qa from the explainx registry; install was straightforward and the SKILL.md answered most questions upfront.
Solid pick for teams standardizing on skills: video-blur-qa is focused, and the summary matches what you get after install.
showing 1-10 of 32