artifact-detect
This skill should be used when the user asks "what's wrong with this video", "why does this video look bad", "detect video artifacts", "find quality issues", "video has artifacts", "identify compression artifacts", or wants to diagnose specific quality problems in a video file.
What this skill does
# Video Artifact Detection
Identify and diagnose specific quality issues in video files. Focus on visual inspection techniques and artifact recognition, not fixing the issues.
## Artifact Categories
### 1. Compression Artifacts
**Blocking (macroblocking)**
- Square or rectangular blocks visible in motion or gradients
- Caused by: Low bitrate encoding, high quantization
- Where to look: Fast motion scenes, areas with subtle color gradients
- Severity indicator: More visible = worse quality
**Banding**
- Visible steps/bands in color gradients instead of smooth transitions
- Caused by: Low bit depth, aggressive quantization, source issues
- Where to look: Sky gradients, dark shadows, smooth surfaces
- Note: May be source issue, not encoding issue
**Mosquito noise**
- Fuzzy, shifting noise around sharp edges
- Caused by: Lossy compression struggling with hard edges
- Where to look: Text overlays, high-contrast edges, animation line art
**DCT ringing**
- Wavy patterns near sharp edges
- Caused by: JPEG-based compression (H.264/H.265 use DCT)
- Where to look: Around text, logos, sharp transitions
### 2. Post-Processing Damage
**Haloing/Ringing from sharpening**
- Bright or dark outlines around edges
- Caused by: Oversharpening filters (unsharp mask, warp sharp)
- Where to look: Around character outlines, text, any hard edges
- Key sign: Consistent glow on one side of lines throughout video
**Line warping**
- Straight lines appear wavy or distorted
- Caused by: Aggressive sharpening algorithms
- Where to look: Architecture, text, grid patterns
**Oversharpened texture**
- Unnatural crispness, "crunchy" appearance
- Caused by: Excessive sharpening combined with compression
- Comparison: Should look natural, not like HDR photo filter
### 3. Upscaling Artifacts
**AI hallucinations**
- Invented details that weren't in source
- Caused by: AI upscaling (Real-ESRGAN, Topaz, etc.)
- Where to look: Fine textures (hair, fabric), text, repeating patterns
- Key sign: Details that seem too sharp or don't match surrounding areas
**Interpolation blur**
- Soft, smeared appearance on edges
- Caused by: Bilinear/bicubic upscaling
- Where to look: Diagonal lines, fine details
- Comparison: Native resolution looks sharper
**Stairstepping (aliasing)**
- Jagged edges on diagonal lines
- Caused by: Poor upscaling algorithm, no anti-aliasing
- Where to look: Diagonal lines, curved edges
### 4. Color Issues
**Wrong color matrix symptoms**
- Overall color shift (too green, too magenta)
- Caused by: BT.601/BT.709 mismatch
- Key sign: Colors look "off" but evenly across entire frame
- Verification: Compare to known-good source
**Chroma shift**
- Color misaligned with luminance
- Caused by: Wrong chroma location, processing error
- Where to look: Hard edges - look for color fringing on one side
- Key sign: Consistent colored glow on same side of all edges
**Double range compression**
- Crushed blacks, blown highlights
- Caused by: Limited range treated as full (or vice versa) multiple times
- Where to look: Dark shadows (crushed to pure black), bright areas (clipped)
### 5. Frame Rate Issues
**Judder/stutter**
- Uneven motion, frames "jumping"
- Caused by: Frame rate mismatch, bad pulldown
- Where to look: Panning shots, steady motion
**Duplicate frames**
- Same frame appears twice (or more)
- Caused by: Frame rate conversion errors
- Detection: Step frame-by-frame, compare adjacent frames
**Telecine combing**
- Horizontal lines across moving objects
- Caused by: Improper deinterlacing of telecined content
- Where to look: Horizontal edges during motion
- Note: Not all combing = interlaced. May be telecined (3:2 pulldown)
## Visual Inspection Checklist
When examining a video for artifacts, inspect these areas in order:
### High-Priority Areas
1. **Dark gradients and shadows** - Reveals banding, blocking, crushed blacks
2. **Strong colors (especially dark reds)** - Shows compression stress, color issues
3. **Around sharp edges** - Reveals haloing, ringing, mosquito noise, chroma shift
4. **On-screen text and logos** - Shows sharpness issues, compression artifacts clearly
5. **Image borders** - Often reveals encoding problems first
### Motion-Specific Checks
6. **Panning shots** - Reveals judder, duplicate frames, motion blur issues
7. **Fast action scenes** - Shows blocking, motion compensation failures
8. **Static grain/texture areas** - Reveals whether grain was filtered or preserved
### Comparison Technique
To spot artifacts effectively:
- View at 100% zoom (1:1 pixel mapping)
- Compare against known-good source if available
- Use frame-by-frame stepping for temporal artifacts
- Toggle between source and processed version rapidly
## Reporting Findings
When reporting detected artifacts, include:
1. **Artifact type** - Name from categories above
2. **Severity** - Subtle, moderate, severe
3. **Location** - Where in frame, what type of content affected
4. **Likely cause** - Based on artifact characteristics
5. **Fixability** - Can be fixed (retag, re-encode from source) or permanent damage
## Tools for Detection
- **mpv**: Best playback, press `i` for technical info, frame step with `.` and `,`
- **vs-preview**: Frame-by-frame comparison with VapourSynth
- **SlowPics**: For sharing comparison screenshots (not imgsli - converts to JPEG)
## Additional Resources
For detailed artifact catalogs with visual examples:
- **`${CLAUDE_PLUGIN_ROOT}/references/artifacts.md`** - Comprehensive artifact guide
- **`${CLAUDE_PLUGIN_ROOT}/references/color-space.md`** - Color issue diagnosis
Related in Image & Video
watch
IncludedWatch a video (URL or local path). Downloads with yt-dlp, extracts auto-scaled frames with ffmpeg, pulls the transcript from captions (or Whisper API fallback), and hands the result to Claude so it can answer questions about what's in the video.
physical-ai-defect-image-generation
IncludedUse when the user wants to orchestrate defect image generation, run associated setup, or handle outputs on OSMO. The Day 0 path handles cold-start with USD-to-ROI, image-edit augmentation, and AnomalyGen to create initial PCBA datasets. The Day 1 path performs inference and labeling on real images. This skill helps with first-time asset setup, creation of finetuning checkpoints, and configuring deployment. Trigger keywords: defect image generation, dig workflow, dig pipeline, defect image detection workflow, aoi pipeline, aoi anomalygen, usd2roi anomalygen, day 0 pcba, day 1 pcba, day 1 real-photo alignment, day 1 manual roi, metal surface anomaly, glass defect, anomalygen finetune, setup_pcb, setup_metal, setup_glass, setup_pretrained, dig setup, dig datasets, dig pretrained checkpoint, dig image-edit endpoint.
accelint-react-best-practices
IncludedReact performance optimization and best practices. ALWAYS use this skill when working with any React code - writing components, hooks, JSX; refactoring; optimizing re-renders, memoization, state management; reviewing for performance; fixing hydration mismatches; debugging infinite re-renders, stale closures, input focus loss, animations restarting; preventing remounting; implementing transitions, lazy initialization, effect dependencies. Even simple React tasks benefit from these patterns. Covers React 19+ (useEffectEvent, Activity, ref props). Triggers - useEffect, useState, useMemo, useCallback, memo, inline components, nested components, components inside components, re-render, performance, hydration, SSR, Next.js, useDeferredValue, combined hooks.
elevenlabs-agents
IncludedBuild conversational AI voice agents with ElevenLabs Platform using React, JavaScript, React Native, or Swift SDKs. Configure agents, tools (client/server/MCP), RAG knowledge bases, multi-voice, and Scribe real-time STT. Use when: building voice chat interfaces, implementing AI phone agents with Twilio, configuring agent workflows or tools, adding RAG knowledge bases, testing with CLI "agents as code", or troubleshooting deprecated @11labs packages, Android audio cutoff, CSP violations, dynamic variables, or WebRTC config. Keywords: ElevenLabs Agents, ElevenLabs voice agents, AI voice agents, conversational AI, @elevenlabs/react, @elevenlabs/client, @elevenlabs/react-native, @elevenlabs/elevenlabs-js, @elevenlabs/agents-cli, elevenlabs SDK, voice AI, TTS, text-to-speech, ASR, speech recognition, turn-taking model, WebRTC voice, WebSocket voice, ElevenLabs conversation, agent system prompt, agent tools, agent knowledge base, RAG voice agents, multi-voice agents, pronunciation dictionary, voice speed control, elevenlabs scribe, @11labs deprecated, Android audio cutoff, CSP violation elevenlabs, dynamic variables elevenlabs, case-sensitive tool names, webhook authentication
humanizer
IncludedHumanize AI-generated text by detecting and removing patterns typical of LLM output. Rewrites text to sound natural, specific, and human. Uses 28 pattern detectors, 560+ AI vocabulary terms across 3 tiers, and statistical analysis (burstiness, type-token ratio, readability) for comprehensive detection. Use when asked to humanize text, de-AI writing, make content sound more natural/human, review writing for AI patterns, score text for AI detection, or improve AI-generated drafts. Covers content, language, style, communication, and filler categories.
generating-mermaid-diagrams
IncludedSalesforce architecture diagrams using Mermaid with ASCII fallback. Use this skill when generating text-based diagrams for Salesforce architecture, OAuth flows, ERDs, integration sequences, or Agentforce structure. TRIGGER when: user says "diagram", "visualize", "ERD", or asks for sequence diagrams, flowcharts, class diagrams, or architecture visualizations in Mermaid. DO NOT TRIGGER when: user wants PNG/SVG image output (use generating-visual-diagrams), or asks about non-Salesforce systems.