{ "schemaVersion": "1.0", "item": { "slug": "deai-image", "name": "Deai Image", "source": "tencent", "type": "skill", "category": "AI 智能", "sourceUrl": "https://clawhub.ai/swaylq/deai-image", "canonicalUrl": "https://clawhub.ai/swaylq/deai-image", "targetPlatform": "OpenClaw" }, "install": { "downloadMode": "redirect", "downloadUrl": "/downloads/deai-image", "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=deai-image", "sourcePlatform": "tencent", "targetPlatform": "OpenClaw", "installMethod": "Manual import", "extraction": "Extract archive", "prerequisites": [ "OpenClaw" ], "packageFormat": "ZIP package", "includedAssets": [ "README.md", "SKILL.md", "package.json", "scripts/check_deps.sh", "scripts/deai.py", "scripts/deai.sh" ], "primaryDoc": "SKILL.md", "quickSetup": [ "Download the package from Yavira.", "Extract the archive and review SKILL.md first.", "Import or place the package into your OpenClaw setup." ], "agentAssist": { "summary": "Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually.", "steps": [ "Download the package from Yavira.", "Extract it into a folder your agent can access.", "Paste one of the prompts below and point your agent at the extracted folder." ], "prompts": [ { "label": "New install", "body": "I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Then review README.md for any prerequisites, environment setup, or post-install checks. Tell me what you changed and call out any manual steps you could not complete." }, { "label": "Upgrade existing", "body": "I downloaded an updated skill package from Yavira. Read SKILL.md from the extracted folder, compare it with my current installation, and upgrade it while preserving any custom configuration unless the package docs explicitly say otherwise. Then review README.md for any prerequisites, environment setup, or post-install checks. Summarize what changed and any follow-up checks I should run." } ] }, "sourceHealth": { "source": "tencent", "status": "healthy", "reason": "direct_download_ok", "recommendedAction": "download", "checkedAt": "2026-04-30T16:55:25.780Z", "expiresAt": "2026-05-07T16:55:25.780Z", "httpStatus": 200, "finalUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=network", "contentType": "application/zip", "probeMethod": "head", "details": { "probeUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=network", "contentDisposition": "attachment; filename=\"network-1.0.0.zip\"", "redirectLocation": null, "bodySnippet": null }, "scope": "source", "summary": "Source download looks usable.", "detail": "Yavira can redirect you to the upstream package for this source.", "primaryActionLabel": "Download for OpenClaw", "primaryActionHref": "/downloads/deai-image" }, "validation": { "installChecklist": [ "Use the Yavira download entry.", "Review SKILL.md after the package is downloaded.", "Confirm the extracted package contains the expected setup assets." ], "postInstallChecks": [ "Confirm the extracted package includes the expected docs or setup files.", "Validate the skill or prompts are available in your target agent workspace.", "Capture any manual follow-up steps the agent could not complete." ] }, "downloadPageUrl": "https://openagent3.xyz/downloads/deai-image", "agentPageUrl": "https://openagent3.xyz/skills/deai-image/agent", "manifestUrl": "https://openagent3.xyz/skills/deai-image/agent.json", "briefUrl": "https://openagent3.xyz/skills/deai-image/agent.md" }, "agentAssist": { "summary": "Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually.", "steps": [ "Download the package from Yavira.", "Extract it into a folder your agent can access.", "Paste one of the prompts below and point your agent at the extracted folder." ], "prompts": [ { "label": "New install", "body": "I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Then review README.md for any prerequisites, environment setup, or post-install checks. Tell me what you changed and call out any manual steps you could not complete." }, { "label": "Upgrade existing", "body": "I downloaded an updated skill package from Yavira. Read SKILL.md from the extracted folder, compare it with my current installation, and upgrade it while preserving any custom configuration unless the package docs explicitly say otherwise. Then review README.md for any prerequisites, environment setup, or post-install checks. Summarize what changed and any follow-up checks I should run." } ] }, "documentation": { "source": "clawhub", "primaryDoc": "SKILL.md", "sections": [ { "title": "AI Image De-Fingerprinting Skill", "body": "Comprehensive CLI for removing AI detection patterns from AI-generated images. Transforms detectable AI images into human-camera-like photographs using multiple processing techniques.\n\nSupported Models: Midjourney, DALL-E 3, Stable Diffusion, Flux, Firefly, Leonardo, and more." }, { "title": "Quick Start", "body": "# Basic processing (medium strength)\npython scripts/deai.py input.png\n\n# Specify output file\npython scripts/deai.py input.png -o output.jpg\n\n# Adjust processing strength\npython scripts/deai.py input.png --strength heavy\n\n# Only strip metadata (fastest)\npython scripts/deai.py input.png --no-metadata\n\n# Batch process directory\npython scripts/deai.py input_dir/ --batch\n\n# Pure Bash version (no Python needed)\nbash scripts/deai.sh input.png output.jpg" }, { "title": "How It Works", "body": "AI-generated images contain multiple detection layers:" }, { "title": "Detection Vectors", "body": "Metadata: EXIF tags revealing generation tool, C2PA watermarks\nFrequency Domain: DCT coefficient patterns unique to diffusion models\nPixel Patterns: Over-smoothness, unnatural noise distribution\nVisual Features: Perfect lighting, repetitive textures" }, { "title": "Processing Pipeline", "body": "Our de-fingerprinting pipeline applies 7 transformation stages:\n\nInput → Metadata Strip → Grain Addition → Color Adjustment → \nBlur/Sharpen → Resize Cycle → JPEG Recompress → Final Metadata Clean → Output\n\nStage Details\n\nStagePurposeTechniqueMetadata StripRemove EXIF/C2PA/JUMBF tagsExifToolGrain AdditionAdd camera sensor noisePoisson/Gaussian noise overlayColor AdjustmentBreak color distribution patternsContrast/saturation/brightness tweakBlur/SharpenDisrupt edge detection patternsGaussian blur + unsharp maskResize CycleIntroduce resampling artifactsDownscale → upscale with LanczosJPEG RecompressAdd compression artifactsQuality 75 → 95 cycleFinal CleanEnsure no metadata leakageExifTool re-run" }, { "title": "Processing Strength", "body": "Choose strength based on detection risk vs quality tradeoff:\n\nStrengthDescriptionSuccess RateQuality LosslightMinimal processing, preserve quality35-45%Very lowmediumBalanced (default)50-65%LowheavyAggressive processing65-80%Medium\n\nSuccess rate = percentage of images passing common AI detectors (Hive, Illuminarty, AI or Not)" }, { "title": "Single Image Processing", "body": "# Default medium strength\npython scripts/deai.py ai_portrait.png\n\n# Light processing for high-quality images\npython scripts/deai.py artwork.png --strength light -o clean_artwork.jpg\n\n# Heavy processing for stubborn detection\npython scripts/deai.py midjourney_out.png --strength heavy" }, { "title": "Batch Processing", "body": "# Process entire directory\npython scripts/deai.py ./ai_images/ --batch -o ./cleaned/\n\n# Batch with specific strength\npython scripts/deai.py ./gallery/*.png --batch --strength heavy" }, { "title": "Metadata-Only Mode", "body": "# Only strip metadata (instant, no quality loss)\npython scripts/deai.py image.jpg --no-metadata" }, { "title": "Using Bash Version", "body": "# No Python/Pillow needed, pure ImageMagick + ExifTool\nbash scripts/deai.sh input.png output.jpg\n\n# Specify strength\nbash scripts/deai.sh input.png output.jpg heavy" }, { "title": "Required", "body": "ImageMagick (7.0+) — Image processing engine\nExifTool — Metadata manipulation\nPython 3.7+ (for deai.py)\nPillow (Python imaging library)\nNumPy (for deai.py)" }, { "title": "Check Installation", "body": "bash scripts/check_deps.sh\n\nThis will verify all dependencies and provide installation commands if missing." }, { "title": "Manual Installation", "body": "Debian/Ubuntu:\n\nsudo apt update\nsudo apt install -y imagemagick libimage-exiftool-perl python3 python3-pip\npip3 install Pillow numpy\n\nmacOS:\n\nbrew install imagemagick exiftool python3\npip3 install Pillow numpy\n\nFedora/RHEL:\n\nsudo dnf install -y ImageMagick perl-Image-ExifTool python3-pip\npip3 install Pillow numpy" }, { "title": "deai.py (Python Version)", "body": "python scripts/deai.py [options]\n\nArguments:\n input Input image file or directory (batch mode)\n\nOptions:\n -o, --output FILE Output file path (default: input_deai.jpg)\n --strength LEVEL Processing strength: light|medium|heavy (default: medium)\n --no-metadata Only strip metadata, skip image processing\n --batch Process entire directory\n -q, --quiet Suppress progress output\n -v, --verbose Show detailed processing steps\n\nExamples:\n python scripts/deai.py image.png\n python scripts/deai.py image.png -o clean.jpg --strength heavy\n python scripts/deai.py folder/ --batch" }, { "title": "deai.sh (Bash Version)", "body": "bash scripts/deai.sh [strength]\n\nArguments:\n input Input image file\n output Output file path\n strength light|medium|heavy (default: medium)\n\nExamples:\n bash scripts/deai.sh input.png output.jpg\n bash scripts/deai.sh input.png output.jpg heavy" }, { "title": "Common AI Detectors", "body": "DetectorMethodBypass RateHive ModerationDeep learning model50-70% (medium)IlluminartyComputer vision analysis60-75% (medium)AI or NotBinary classification55-70% (medium)SynthIDPixel-level watermark35-50% (heavy)C2PA VerifyMetadata check100% (metadata strip)" }, { "title": "What This Skill Cannot Do", "body": "❌ Not a Silver Bullet:\n\nCannot guarantee 100% bypass of all detectors\nAdvanced detectors (SynthID) require more aggressive processing\nNew detection methods may emerge\n\n❌ Limitations:\n\nProcessing reduces image quality (tradeoff necessary)\nSome detectors use multiple layers (metadata + pixel + frequency)\nExtremely aggressive processing may introduce visible artifacts\n\n✅ What It DOES Do:\n\nSignificantly reduces detection probability (40-80%)\nRemoves metadata watermarks (100% effective)\nMaintains reasonable visual quality\nBatch processes entire collections" }, { "title": "Verification Workflow", "body": "Process Image:\npython scripts/deai.py ai_image.png -o clean.jpg --strength medium\n\n\n\nTest on Multiple Detectors:\n\nHive Moderation\nIlluminarty\nAI or Not\n\n\n\nIf Still Detected:\n\nIncrease strength: --strength heavy\nTry multiple passes\nManual touch-ups (add slight noise in photo editor)\n\n\n\nQuality Check:\n\nCompare original vs processed\nEnsure no visible artifacts\nVerify colors/details preserved" }, { "title": "Custom Processing Pipeline", "body": "Edit scripts/deai.py to adjust parameters:\n\n# Noise strength (line ~80)\nnoise = np.random.normal(0, 3, img_array.shape) # Increase 3 → 5 for more grain\n\n# Contrast adjustment (line ~95)\nenhancer.enhance(1.05) # Increase 1.05 → 1.08 for stronger effect\n\n# JPEG quality (line ~120)\nimg.save(temp_path, \"JPEG\", quality=80) # Decrease 80 → 70 for more compression" }, { "title": "Combining with External Tools", "body": "# Step 1: De-fingerprint\npython scripts/deai.py ai_gen.png -o step1.jpg\n\n# Step 2: Add subtle texture overlay (GIMP/Photoshop)\n# (Manual step)\n\n# Step 3: Re-strip metadata\nexiftool -all= step1_edited.jpg" }, { "title": "For Social Media", "body": "Use medium strength (good balance)\nOutput as JPEG (universal compatibility)\nTest on platform's upload flow before posting" }, { "title": "For Professional Use", "body": "Start with light (preserve quality)\nManual review each output\nKeep originals in secure storage\nDocument processing steps" }, { "title": "For Research/Testing", "body": "Use heavy for stress testing\nCompare multiple detectors\nDocument success/failure patterns" }, { "title": "Legal & Ethical Notice", "body": "⚠️ Use Responsibly:\n\nThis tool is intended for:\n\n✅ Personal creative projects\n✅ Academic research on AI detection\n✅ Security testing (authorized)\n✅ Understanding detection mechanisms\n\nDO NOT use for:\n\n❌ Fraud or deception\n❌ Impersonating human creators\n❌ Bypassing platform policies without authorization\n❌ Creating misleading content\n\nLegal Risks:\n\nSome jurisdictions (e.g., COPIED Act 2024) may restrict watermark removal\nPlatform terms of service often prohibit AI content masking\nCommercial use may have additional legal requirements\n\nYou are responsible for compliance with applicable laws and terms of service." }, { "title": "\"Command not found: exiftool\"", "body": "# Install ExifTool\nsudo apt install libimage-exiftool-perl # Debian/Ubuntu\nbrew install exiftool # macOS" }, { "title": "\"ImportError: No module named PIL\"", "body": "pip3 install Pillow numpy" }, { "title": "\"ImageMagick policy.xml blocks operation\"", "body": "# Edit /etc/ImageMagick-7/policy.xml\n# Change: \n# To: " }, { "title": "Processing is slow on large images", "body": "# Pre-resize before processing\nmagick large.png -resize 2048x2048\\> resized.png\npython scripts/deai.py resized.png" }, { "title": "Output looks too grainy/noisy", "body": "# Use light strength\npython scripts/deai.py input.png --strength light" }, { "title": "Running Tests", "body": "# Test dependency check\nbash scripts/check_deps.sh\n\n# Test single image (verbose)\npython scripts/deai.py test_images/sample.png -v\n\n# Test batch mode\nmkdir test_output\npython scripts/deai.py test_images/ --batch -o test_output/" }, { "title": "Contributing", "body": "Improvements welcome! Focus areas:\n\nNew detection bypass techniques\nQuality preservation algorithms\nSupport for more image formats (HEIC, AVIF)\nIntegration with detection APIs" }, { "title": "References", "body": "Detection Research:\n\nHu, Y., et al. (2024). \"Stable signature is unstable: Removing image watermark from diffusion models.\" arXiv:2405.07145\nIEEE Spectrum: UnMarker tool analysis\n\nOpen Source Projects:\n\nSynthid-Bypass — ComfyUI watermark removal\nC2PAC — C2PA metadata tools\n\nDetection Tools:\n\nHive Moderation\nContent Credentials Verify\nGoogle SynthID\n\nVersion: 1.0.0\nLicense: MIT (for educational/research use)\nMaintainer: voidborne-d\nLast Updated: 2026-02-23" } ], "body": "AI Image De-Fingerprinting Skill\n\nComprehensive CLI for removing AI detection patterns from AI-generated images. Transforms detectable AI images into human-camera-like photographs using multiple processing techniques.\n\nSupported Models: Midjourney, DALL-E 3, Stable Diffusion, Flux, Firefly, Leonardo, and more.\n\nQuick Start\n# Basic processing (medium strength)\npython scripts/deai.py input.png\n\n# Specify output file\npython scripts/deai.py input.png -o output.jpg\n\n# Adjust processing strength\npython scripts/deai.py input.png --strength heavy\n\n# Only strip metadata (fastest)\npython scripts/deai.py input.png --no-metadata\n\n# Batch process directory\npython scripts/deai.py input_dir/ --batch\n\n# Pure Bash version (no Python needed)\nbash scripts/deai.sh input.png output.jpg\n\nHow It Works\n\nAI-generated images contain multiple detection layers:\n\nDetection Vectors\nMetadata: EXIF tags revealing generation tool, C2PA watermarks\nFrequency Domain: DCT coefficient patterns unique to diffusion models\nPixel Patterns: Over-smoothness, unnatural noise distribution\nVisual Features: Perfect lighting, repetitive textures\nProcessing Pipeline\n\nOur de-fingerprinting pipeline applies 7 transformation stages:\n\nInput → Metadata Strip → Grain Addition → Color Adjustment → \nBlur/Sharpen → Resize Cycle → JPEG Recompress → Final Metadata Clean → Output\n\nStage Details\nStage\tPurpose\tTechnique\nMetadata Strip\tRemove EXIF/C2PA/JUMBF tags\tExifTool\nGrain Addition\tAdd camera sensor noise\tPoisson/Gaussian noise overlay\nColor Adjustment\tBreak color distribution patterns\tContrast/saturation/brightness tweak\nBlur/Sharpen\tDisrupt edge detection patterns\tGaussian blur + unsharp mask\nResize Cycle\tIntroduce resampling artifacts\tDownscale → upscale with Lanczos\nJPEG Recompress\tAdd compression artifacts\tQuality 75 → 95 cycle\nFinal Clean\tEnsure no metadata leakage\tExifTool re-run\nProcessing Strength\n\nChoose strength based on detection risk vs quality tradeoff:\n\nStrength\tDescription\tSuccess Rate\tQuality Loss\nlight\tMinimal processing, preserve quality\t35-45%\tVery low\nmedium\tBalanced (default)\t50-65%\tLow\nheavy\tAggressive processing\t65-80%\tMedium\n\nSuccess rate = percentage of images passing common AI detectors (Hive, Illuminarty, AI or Not)\n\nUsage Examples\nSingle Image Processing\n# Default medium strength\npython scripts/deai.py ai_portrait.png\n\n# Light processing for high-quality images\npython scripts/deai.py artwork.png --strength light -o clean_artwork.jpg\n\n# Heavy processing for stubborn detection\npython scripts/deai.py midjourney_out.png --strength heavy\n\nBatch Processing\n# Process entire directory\npython scripts/deai.py ./ai_images/ --batch -o ./cleaned/\n\n# Batch with specific strength\npython scripts/deai.py ./gallery/*.png --batch --strength heavy\n\nMetadata-Only Mode\n# Only strip metadata (instant, no quality loss)\npython scripts/deai.py image.jpg --no-metadata\n\nUsing Bash Version\n# No Python/Pillow needed, pure ImageMagick + ExifTool\nbash scripts/deai.sh input.png output.jpg\n\n# Specify strength\nbash scripts/deai.sh input.png output.jpg heavy\n\nDependencies\nRequired\nImageMagick (7.0+) — Image processing engine\nExifTool — Metadata manipulation\nPython 3.7+ (for deai.py)\nPillow (Python imaging library)\nNumPy (for deai.py)\nCheck Installation\nbash scripts/check_deps.sh\n\n\nThis will verify all dependencies and provide installation commands if missing.\n\nManual Installation\n\nDebian/Ubuntu:\n\nsudo apt update\nsudo apt install -y imagemagick libimage-exiftool-perl python3 python3-pip\npip3 install Pillow numpy\n\n\nmacOS:\n\nbrew install imagemagick exiftool python3\npip3 install Pillow numpy\n\n\nFedora/RHEL:\n\nsudo dnf install -y ImageMagick perl-Image-ExifTool python3-pip\npip3 install Pillow numpy\n\nCommand Reference\ndeai.py (Python Version)\npython scripts/deai.py [options]\n\nArguments:\n input Input image file or directory (batch mode)\n\nOptions:\n -o, --output FILE Output file path (default: input_deai.jpg)\n --strength LEVEL Processing strength: light|medium|heavy (default: medium)\n --no-metadata Only strip metadata, skip image processing\n --batch Process entire directory\n -q, --quiet Suppress progress output\n -v, --verbose Show detailed processing steps\n\nExamples:\n python scripts/deai.py image.png\n python scripts/deai.py image.png -o clean.jpg --strength heavy\n python scripts/deai.py folder/ --batch\n\ndeai.sh (Bash Version)\nbash scripts/deai.sh [strength]\n\nArguments:\n input Input image file\n output Output file path\n strength light|medium|heavy (default: medium)\n\nExamples:\n bash scripts/deai.sh input.png output.jpg\n bash scripts/deai.sh input.png output.jpg heavy\n\nUnderstanding Detection\nCommon AI Detectors\nDetector\tMethod\tBypass Rate\nHive Moderation\tDeep learning model\t50-70% (medium)\nIlluminarty\tComputer vision analysis\t60-75% (medium)\nAI or Not\tBinary classification\t55-70% (medium)\nSynthID\tPixel-level watermark\t35-50% (heavy)\nC2PA Verify\tMetadata check\t100% (metadata strip)\nWhat This Skill Cannot Do\n\n❌ Not a Silver Bullet:\n\nCannot guarantee 100% bypass of all detectors\nAdvanced detectors (SynthID) require more aggressive processing\nNew detection methods may emerge\n\n❌ Limitations:\n\nProcessing reduces image quality (tradeoff necessary)\nSome detectors use multiple layers (metadata + pixel + frequency)\nExtremely aggressive processing may introduce visible artifacts\n\n✅ What It DOES Do:\n\nSignificantly reduces detection probability (40-80%)\nRemoves metadata watermarks (100% effective)\nMaintains reasonable visual quality\nBatch processes entire collections\nVerification Workflow\n\nProcess Image:\n\npython scripts/deai.py ai_image.png -o clean.jpg --strength medium\n\n\nTest on Multiple Detectors:\n\nHive Moderation\nIlluminarty\nAI or Not\n\nIf Still Detected:\n\nIncrease strength: --strength heavy\nTry multiple passes\nManual touch-ups (add slight noise in photo editor)\n\nQuality Check:\n\nCompare original vs processed\nEnsure no visible artifacts\nVerify colors/details preserved\nAdvanced Usage\nCustom Processing Pipeline\n\nEdit scripts/deai.py to adjust parameters:\n\n# Noise strength (line ~80)\nnoise = np.random.normal(0, 3, img_array.shape) # Increase 3 → 5 for more grain\n\n# Contrast adjustment (line ~95)\nenhancer.enhance(1.05) # Increase 1.05 → 1.08 for stronger effect\n\n# JPEG quality (line ~120)\nimg.save(temp_path, \"JPEG\", quality=80) # Decrease 80 → 70 for more compression\n\nCombining with External Tools\n# Step 1: De-fingerprint\npython scripts/deai.py ai_gen.png -o step1.jpg\n\n# Step 2: Add subtle texture overlay (GIMP/Photoshop)\n# (Manual step)\n\n# Step 3: Re-strip metadata\nexiftool -all= step1_edited.jpg\n\nBest Practices\nFor Social Media\nUse medium strength (good balance)\nOutput as JPEG (universal compatibility)\nTest on platform's upload flow before posting\nFor Professional Use\nStart with light (preserve quality)\nManual review each output\nKeep originals in secure storage\nDocument processing steps\nFor Research/Testing\nUse heavy for stress testing\nCompare multiple detectors\nDocument success/failure patterns\nLegal & Ethical Notice\n\n⚠️ Use Responsibly:\n\nThis tool is intended for:\n\n✅ Personal creative projects\n✅ Academic research on AI detection\n✅ Security testing (authorized)\n✅ Understanding detection mechanisms\n\nDO NOT use for:\n\n❌ Fraud or deception\n❌ Impersonating human creators\n❌ Bypassing platform policies without authorization\n❌ Creating misleading content\n\nLegal Risks:\n\nSome jurisdictions (e.g., COPIED Act 2024) may restrict watermark removal\nPlatform terms of service often prohibit AI content masking\nCommercial use may have additional legal requirements\n\nYou are responsible for compliance with applicable laws and terms of service.\n\nTroubleshooting\n\"Command not found: exiftool\"\n# Install ExifTool\nsudo apt install libimage-exiftool-perl # Debian/Ubuntu\nbrew install exiftool # macOS\n\n\"ImportError: No module named PIL\"\npip3 install Pillow numpy\n\n\"ImageMagick policy.xml blocks operation\"\n# Edit /etc/ImageMagick-7/policy.xml\n# Change: \n# To: \n\nProcessing is slow on large images\n# Pre-resize before processing\nmagick large.png -resize 2048x2048\\> resized.png\npython scripts/deai.py resized.png\n\nOutput looks too grainy/noisy\n# Use light strength\npython scripts/deai.py input.png --strength light\n\nDevelopment\nRunning Tests\n# Test dependency check\nbash scripts/check_deps.sh\n\n# Test single image (verbose)\npython scripts/deai.py test_images/sample.png -v\n\n# Test batch mode\nmkdir test_output\npython scripts/deai.py test_images/ --batch -o test_output/\n\nContributing\n\nImprovements welcome! Focus areas:\n\nNew detection bypass techniques\nQuality preservation algorithms\nSupport for more image formats (HEIC, AVIF)\nIntegration with detection APIs\nReferences\n\nDetection Research:\n\nHu, Y., et al. (2024). \"Stable signature is unstable: Removing image watermark from diffusion models.\" arXiv:2405.07145\nIEEE Spectrum: UnMarker tool analysis\n\nOpen Source Projects:\n\nSynthid-Bypass — ComfyUI watermark removal\nC2PAC — C2PA metadata tools\n\nDetection Tools:\n\nHive Moderation\nContent Credentials Verify\nGoogle SynthID\n\nVersion: 1.0.0\nLicense: MIT (for educational/research use)\nMaintainer: voidborne-d\nLast Updated: 2026-02-23" }, "trust": { "sourceLabel": "tencent", "provenanceUrl": "https://clawhub.ai/swaylq/deai-image", "publisherUrl": "https://clawhub.ai/swaylq/deai-image", "owner": "swaylq", "version": "1.0.0", "license": null, "verificationStatus": "Indexed source record" }, "links": { "detailUrl": "https://openagent3.xyz/skills/deai-image", "downloadUrl": "https://openagent3.xyz/downloads/deai-image", "agentUrl": "https://openagent3.xyz/skills/deai-image/agent", "manifestUrl": "https://openagent3.xyz/skills/deai-image/agent.json", "briefUrl": "https://openagent3.xyz/skills/deai-image/agent.md" } }