[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7BVJC3PSEQDMM67Y35K6718":3,"skill-same:01M7BVJC3PSEQDMM67Y35K6718":45},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":24,"commitSha":25,"skillDirectory":26,"forksCount":27,"repositoryHtmlUrl":28,"repositoryHomepageUrl":6,"ownerHtmlUrl":29,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":30,"install":31,"resources":44},"01M7BVJC3PSEQDMM67Y35K6718","image-ocr",null,"Extract text content from images using Tesseract OCR via Python","linyouxue\u002Fskill-repair-benchmark","linyouxue","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F216151464?v=4","01M7BJEVCXKD74KH6FG9CBVSQB","林宥学","01M7BJEVE02NP217V93BKZFC0J",0,"Apache-2.0","https:\u002F\u002Fgithub.com\u002Flinyouxue\u002Fskill-repair-benchmark\u002Fblob\u002Fmain\u002Fevaluation\u002Fresults\u002FMMG2Skill-Opus47-Gold15-20260924\u002Fruns\u002Fmmg2skill-opus47-gold15-20260924\u002Fjpg-ocr-stat-mmg2skill-opus47-r001\u002Finputs\u002Fskills\u002Fimage-ocr\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: image-ocr\ndescription: Extract text content from images using Tesseract OCR via Python\n---\n\n# Image OCR Skill\n## Purpose\nThis skill enables accurate text extraction from image files (JPG, PNG, etc.) using Tesseract OCR via the `pytesseract` Python library. It is suitable for scanned documents, screenshots, photos of text, receipts, forms, and other visual content containing text.\n## When to Use\n- Extracting text from scanned documents or photos\n- Reading text from screenshots or image captures\n- Processing batch image files that contain textual information\n- Converting visual documents to machine-readable text\n- Extracting structured data from forms, receipts, or tables in images\n## Required Libraries\n```python\nimport pytesseract\nfrom PIL import Image, ImageFilter, ImageOps\nimport re\nimport os\n```\n## Critical Workflow Rules\n### Always run the script after editing it\nAfter every edit to the extraction script, immediately re-run `python3 extract.py` and verify the on-disk xlsx before considering the task done. The xlsx must reflect the latest code. Do not end the session on an edit — end on a successful run whose output you have inspected.\n### Preprocessing variants for receipts\nFor OCR of receipts, use multiple preprocessing variants and aggregate results. A robust default set (in order):\n1. Original (default PSM)\n2. Grayscale + autocontrast, `--psm 6`\n3. 2x upscale grayscale + autocontrast, `--psm 6`\n4. 3x upscale grayscale + autocontrast + sharpen, `--psm 4`\nThe 3x upscale + psm 4 variant is specifically effective for fragmented \"Grand Total\" lines that other variants miss.\n```python\ndef ocr_variants(image_path):\n    img = Image.open(image_path)\n    gray = ImageOps.autocontrast(ImageOps.grayscale(img))\n    up2 = gray.resize((gray.width*2, gray.height*2), Image.LANCZOS)\n    up3 = gray.resize((gray.width*3, gray.height*3), Image.LANCZOS).filter(ImageFilter.SHARPEN)\n    return [\n        pytesseract.image_to_string(img),\n        pytesseract.image_to_string(gray, config='--psm 6'),\n        pytesseract.image_to_string(up2, config='--psm 6'),\n        pytesseract.image_to_string(up3, config='--psm 4'),\n    ]\n```\n## Receipt Total Extraction Strategy\n### Keyword tiers (most-specific first)\n1. `GRAND TOTAL`\n2. `TOTAL RM`, `TOTAL: RM`\n3. `TOTAL AMOUNT`\n4. `TOTAL`, `AMOUNT`, `TOTAL DUE`, `AMOUNT DUE`, `BALANCE DUE`, `NETT TOTAL`, `NET TOTAL`\n### Exclusion patterns (skip these lines entirely)\n`SUBTOTAL`, `SUB TOTAL`, `TAX`, `GST`, `SST`, `DISCOUNT`, `CHANGE`, `CASH TENDERED`, `CASH RECEIVED`, `CASH PAYMENT`, `EXCLUDING`, `TOTAL GST`, `TOTAL TAX`, `TOTAL SST`, `TOTAL ITEMS`, `TOTAL QTY`, `TOTAL SAVINGS`, `ROUNDED`, `ROUNDING`, `DESCRIPTION`, table header patterns like `QTY ... PRICE ... AMOUNT`, and payment card names like `VISA`, `MASTERCARD`.\nCritically, `ROUNDED`\u002F`ROUNDING` MUST be excluded — lines such as \"Total Amt Rounded 702.00\" are OCR-garbled rounding lines, not the true total, and will beat the true total in voting if not excluded.\n### Number parsing\n- Strict 2-decimal regex: `r'(\\d{1,3}(?:,\\d{3})*|\\d+)\\.\\d{2}\\b'`\n- Strip commas before converting to float.\n- On a keyword line, take the **last** matching number.\n- Two-line fallback: only when the keyword line itself has **no digits**, look at the next non-empty line and take its last number.\n### Aggregating across variants\n1. Collect (tier, value) candidates from all variants.\n2. Pick the lowest tier index present.\n3. Within that tier, take the majority value; on ties, prefer the value produced by the highest-index variant (upscaled variants are more trustworthy than the default pass).\n## Date Extraction\nHandle these formats and normalize to `YYYY-MM-DD`:\n- `DD\u002FMM\u002FYYYY` or `DD-MM-YYYY`\n- `DD\u002FMM\u002FYY` → prepend `20`\n- `YYYY\u002FMM\u002FDD` or `YYYY-MM-DD`\n## Steps\n1. Enumerate image files under the input directory (sorted).\n2. For each image, run all 4 OCR variants above.\n3. From each variant's text, apply keyword-tier matching with exclusions, using strict 2-decimal regex and the digit-less two-line fallback.\n4. Aggregate candidates across variants using the tier + majority + upscale-preference rule.\n5. Extract date using the multi-format normalizer.\n6. Write results to xlsx (see `xlsx` skill), sorted by filename, one sheet `results`, columns `filename,date,total_amount`. Missing extractions become `None`.\n7. **Run `python3 extract.py`** and inspect the printed rows plus the saved xlsx before finishing.\n## Expected Result\nAn xlsx at the specified path with exactly one sheet `results` and three columns, one row per input image, ordered by filename. Every row reflects the most recent script execution.\n","# Image OCR Skill\n## Purpose\nThis skill enables accurate text extraction from image files (JPG, PNG, etc.) using Tesseract OCR via the `pytesseract` Python library. It is suitable for scanned documents, screenshots, photos of text, receipts, forms, and other visual content containing text.\n## When to Use\n- Extracting text from scanned documents or photos\n- Reading text from screenshots or image captures\n- Processing batch image files that contain textual information\n- Converting visual documents to machine-readable text\n- Extracting structured data from forms, receipts, or tables in images\n## Required Libraries\n```python\nimport pytesseract\nfrom PIL import Image, ImageFilter, ImageOps\nimport re\nimport os\n```\n## Critical Workflow Rules\n### Always run the script after editing it\nAfter every edit to the extraction script, immediately re-run `python3 extract.py` and verify the on-disk xlsx before considering the task done. The xlsx must reflect the latest code. Do not end the session on an edit — end on a successful run whose output you have inspected.\n### Preprocessing variants for receipts\nFor OCR of receipts, use multiple preprocessing variants and aggregate results. A robust default set (in order):\n1. Original (default PSM)\n2. Grayscale + autocontrast, `--psm 6`\n3. 2x upscale grayscale + autocontrast, `--psm 6`\n4. 3x upscale grayscale + autocontrast + sharpen, `--psm 4`\nThe 3x upscale + psm 4 variant is specifically effective for fragmented \"Grand Total\" lines that other variants miss.\n```python\ndef ocr_variants(image_path):\n    img = Image.open(image_path)\n    gray = ImageOps.autocontrast(ImageOps.grayscale(img))\n    up2 = gray.resize((gray.width*2, gray.height*2), Image.LANCZOS)\n    up3 = gray.resize((gray.width*3, gray.height*3), Image.LANCZOS).filter(ImageFilter.SHARPEN)\n    return [\n        pytesseract.image_to_string(img),\n        pytesseract.image_to_string(gray, config='--psm 6'),\n        pytesseract.image_to_string(up2, config='--psm 6'),\n        pytesseract.image_to_string(up3, config='--psm 4'),\n    ]\n```\n## Receipt Total Extraction Strategy\n### Keyword tiers (most-specific first)\n1. `GRAND TOTAL`\n2. `TOTAL RM`, `TOTAL: RM`\n3. `TOTAL AMOUNT`\n4. `TOTAL`, `AMOUNT`, `TOTAL DUE`, `AMOUNT DUE`, `BALANCE DUE`, `NETT TOTAL`, `NET TOTAL`\n### Exclusion patterns (skip these lines entirely)\n`SUBTOTAL`, `SUB TOTAL`, `TAX`, `GST`, `SST`, `DISCOUNT`, `CHANGE`, `CASH TENDERED`, `CASH RECEIVED`, `CASH PAYMENT`, `EXCLUDING`, `TOTAL GST`, `TOTAL TAX`, `TOTAL SST`, `TOTAL ITEMS`, `TOTAL QTY`, `TOTAL SAVINGS`, `ROUNDED`, `ROUNDING`, `DESCRIPTION`, table header patterns like `QTY ... PRICE ... AMOUNT`, and payment card names like `VISA`, `MASTERCARD`.\nCritically, `ROUNDED`\u002F`ROUNDING` MUST be excluded — lines such as \"Total Amt Rounded 702.00\" are OCR-garbled rounding lines, not the true total, and will beat the true total in voting if not excluded.\n### Number parsing\n- Strict 2-decimal regex: `r'(\\d{1,3}(?:,\\d{3})*|\\d+)\\.\\d{2}\\b'`\n- Strip commas before converting to float.\n- On a keyword line, take the **last** matching number.\n- Two-line fallback: only when the keyword line itself has **no digits**, look at the next non-empty line and take its last number.\n### Aggregating across variants\n1. Collect (tier, value) candidates from all variants.\n2. Pick the lowest tier index present.\n3. Within that tier, take the majority value; on ties, prefer the value produced by the highest-index variant (upscaled variants are more trustworthy than the default pass).\n## Date Extraction\nHandle these formats and normalize to `YYYY-MM-DD`:\n- `DD\u002FMM\u002FYYYY` or `DD-MM-YYYY`\n- `DD\u002FMM\u002FYY` → prepend `20`\n- `YYYY\u002FMM\u002FDD` or `YYYY-MM-DD`\n## Steps\n1. Enumerate image files under the input directory (sorted).\n2. For each image, run all 4 OCR variants above.\n3. From each variant's text, apply keyword-tier matching with exclusions, using strict 2-decimal regex and the digit-less two-line fallback.\n4. Aggregate candidates across variants using the tier + majority + upscale-preference rule.\n5. Extract date using the multi-format normalizer.\n6. Write results to xlsx (see `xlsx` skill), sorted by filename, one sheet `results`, columns `filename,date,total_amount`. Missing extractions become `None`.\n7. **Run `python3 extract.py`** and inspect the printed rows plus the saved xlsx before finishing.\n## Expected Result\nAn xlsx at the specified path with exactly one sheet `results` and three columns, one row per input image, ordered by filename. Every row reflects the most recent script execution.\n",{"name":5,"description":7},"main","cdcb6d230f3b15c3c33f8a2bdd16439f43602cbe","evaluation\u002Fresults\u002FMMG2Skill-Opus47-Gold15-20260924\u002Fruns\u002Fmmg2skill-opus47-gold15-20260924\u002Fjpg-ocr-stat-mmg2skill-opus47-r001\u002Finputs\u002Fskills\u002Fimage-ocr",1,"https:\u002F\u002Fgithub.com\u002Flinyouxue\u002Fskill-repair-benchmark","https:\u002F\u002Fgithub.com\u002Flinyouxue","2026-11-14 20:28:35",{"source":8,"skillName":5,"commands":32,"prompt":42,"zipAvailable":43},[33,36,39],{"tool":34,"command":35},"npx","npx skills add linyouxue\u002Fskill-repair-benchmark --skill image-ocr",{"tool":37,"command":38},"bunx","bunx skills add linyouxue\u002Fskill-repair-benchmark --skill image-ocr",{"tool":40,"command":41},"pnpm","pnpm dlx skills add linyouxue\u002Fskill-repair-benchmark --skill image-ocr","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Flinyouxue\u002Fskill-repair-benchmark\u002Fblob\u002Fmain\u002Fevaluation\u002Fresults\u002FMMG2Skill-Opus47-Gold15-20260924\u002Fruns\u002Fmmg2skill-opus47-gold15-20260924\u002Fjpg-ocr-stat-mmg2skill-opus47-r001\u002Finputs\u002Fskills\u002Fimage-ocr\u002FSKILL.md",true,[],[46,53,60,66,71,76],{"skillNo":47,"name":48,"nameZh":6,"description":49,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":50,"categories":51,"keywords":52,"publishedAt":19},"01M7E1HGVRY73NBJ022CJG13X6","video-frame-extraction","Extract frames from video files and save them as images using OpenCV","https:\u002F\u002Fgithub.com\u002Flinyouxue\u002Fskill-repair-benchmark\u002Fblob\u002Fmain\u002Fjpg-ocr-stat\u002Fresult\u002Fskilltta-claude4.7\u002Fgroup_03\u002Frun_03\u002Finputs\u002Fskills\u002Fvideo-frame-extraction\u002FSKILL.md",[],[],{"skillNo":54,"name":55,"nameZh":6,"description":56,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":57,"categories":58,"keywords":59,"publishedAt":19},"01M7E1HAG5R7CJ9C8NYJS771MB","openai-vision","Analyze images and multi-frame sequences using OpenAI GPT vision models","https:\u002F\u002Fgithub.com\u002Flinyouxue\u002Fskill-repair-benchmark\u002Fblob\u002Fmain\u002Fjpg-ocr-stat\u002Fresult\u002Fskilltta-claude4.7\u002Fgroup_03\u002Frun_03\u002Finputs\u002Fskills\u002Fopenai-vision\u002FSKILL.md",[],[],{"skillNo":61,"name":5,"nameZh":6,"description":62,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":63,"categories":64,"keywords":65,"publishedAt":19},"01M7E1H7DADKB6QBWKC2KQ63SZ","Extract text content from images using Tesseract OCR via Python, including structured field extraction from receipts and forms","https:\u002F\u002Fgithub.com\u002Flinyouxue\u002Fskill-repair-benchmark\u002Fblob\u002Fmain\u002Fjpg-ocr-stat\u002Fresult\u002Fskilltta-claude4.7\u002Fgroup_03\u002Frun_03\u002Finputs\u002Fskills\u002Fimage-ocr\u002FSKILL.md",[],[],{"skillNo":67,"name":48,"nameZh":6,"description":49,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":68,"categories":69,"keywords":70,"publishedAt":19},"01M7E1H0WBNT43TENGQY7AVZHD","https:\u002F\u002Fgithub.com\u002Flinyouxue\u002Fskill-repair-benchmark\u002Fblob\u002Fmain\u002Fjpg-ocr-stat\u002Fresult\u002Fskilltta-claude4.7\u002Fgroup_03\u002Frun_02\u002Finputs\u002Fskills\u002Fvideo-frame-extraction\u002FSKILL.md",[],[],{"skillNo":72,"name":55,"nameZh":6,"description":56,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":73,"categories":74,"keywords":75,"publishedAt":19},"01M7E1GTC7D4N4Q2EF03875J9E","https:\u002F\u002Fgithub.com\u002Flinyouxue\u002Fskill-repair-benchmark\u002Fblob\u002Fmain\u002Fjpg-ocr-stat\u002Fresult\u002Fskilltta-claude4.7\u002Fgroup_03\u002Frun_02\u002Finputs\u002Fskills\u002Fopenai-vision\u002FSKILL.md",[],[],{"skillNo":77,"name":5,"nameZh":6,"description":62,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":78,"categories":79,"keywords":80,"publishedAt":19},"01M7E1GQ92CWFMGH53KSSE4WER","https:\u002F\u002Fgithub.com\u002Flinyouxue\u002Fskill-repair-benchmark\u002Fblob\u002Fmain\u002Fjpg-ocr-stat\u002Fresult\u002Fskilltta-claude4.7\u002Fgroup_03\u002Frun_02\u002Finputs\u002Fskills\u002Fimage-ocr\u002FSKILL.md",[],[]]