[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7BQKHYWPVBQSJDADE9TK1P6":3,"skill-same:01M7BQKHYWPVBQSJDADE9TK1P6":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},"01M7BQKHYWPVBQSJDADE9TK1P6","scip-opt",null,"SCIP optimization with PySCIPOpt. Use when facing an optimization problem with an objective, hard constraints, soft penalties, integer decisions, routing, assignment, scheduling, allocation, packing, capacity, inventory, or service-level rules. Prefer modeling and solving the problem with PySCIPOpt when it is available.","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\u002Fdata\u002Fskillsbench87\u002Freferences\u002Fbike-rebalance\u002Fskills\u002Fscip-opt\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: scip-opt\ndescription: SCIP optimization with PySCIPOpt. Use when facing an optimization problem with an objective, hard constraints, soft penalties, integer decisions, routing, assignment, scheduling, allocation, packing, capacity, inventory, or service-level rules. Prefer modeling and solving the problem with PySCIPOpt when it is available.\n---\n\n# SCIP Optimization\n\nUse SCIP through `pyscipopt` when a task asks you to minimize or maximize an objective subject to constraints.\n\nSCIP is a strong open-source optimization solver with a Python API. It is well suited for mixed-integer optimization, routing-style models, assignment models, capacity planning, inventory movement, scheduling, and problems with soft penalties. For benchmark tasks, a SCIP-backed model plus an independent validator is usually safer than a greedy construction.\n\n## When To Use\n\nConsider PySCIPOpt when the request includes:\n\n- an objective such as minimizing cost, distance, time, unmet demand, or penalty;\n- yes\u002Fno choices, route arcs, assignments, selected items, or ordering decisions;\n- integer or continuous quantities such as load, inventory, flow, served units, or slack;\n- hard rules that every valid answer must satisfy;\n- soft rules that can be violated with an explicit penalty.\n\nDo not start by installing another optimization package. First check whether PySCIPOpt is already available:\n\n```python\ntry:\n    from pyscipopt import Model, quicksum\nexcept ImportError as exc:\n    raise RuntimeError(\"PySCIPOpt is required for this optimization approach\") from exc\n```\n\n## Modeling Workflow\n\n1. Identify sets and indices.\n   - Examples: vehicles `K`, stations `N`, jobs `J`, periods `T`, arcs `A`.\n   - Build explicit mappings when input IDs are not contiguous.\n\n2. Define decision variables.\n   - Binary variables for choices, visits, assignments, route arcs, or modes.\n   - Integer variables for counts, loads, inventory moves, or unmet units.\n   - Continuous variables for flows, costs, times, slacks, or resource levels.\n\n3. Add hard constraints.\n   - Conservation, capacity, bounds, linking, continuity, inventory limits, and mutual exclusion.\n\n4. Add soft constraints with explicit slack variables.\n   - Never use Python `abs()` on solver expressions.\n   - Linearize absolute deviation with two inequalities.\n\n5. Set a single objective.\n   - Keep named objective components such as travel cost and penalty cost.\n\n6. Solve with time and gap limits, then apply the incumbent-quality gate below.\n   - Require at least one incumbent to establish feasibility.\n   - Do not extract a final answer until the task's objective-quality criterion is met.\n\n7. Reconstruct and independently validate the output.\n   - Recompute objective components and every hard rule from the reported answer.\n\n## Minimal PySCIPOpt Template\n\n```python\nfrom pyscipopt import Model, quicksum\n\nmodel = Model(\"optimization_model\")\nmodel.hideOutput()\n\nI = range(n_items)\n\nx = {i: model.addVar(vtype=\"B\", name=f\"x_{i}\") for i in I}\namount = {\n    i: model.addVar(vtype=\"I\", lb=0, ub=capacity[i], name=f\"amount_{i}\")\n    for i in I\n}\ndev = {i: model.addVar(lb=0, name=f\"dev_{i}\") for i in I}\n\nfor i in I:\n    model.addCons(amount[i] \u003C= capacity[i] * x[i])\n    model.addCons(amount[i] - target[i] \u003C= dev[i])\n    model.addCons(target[i] - amount[i] \u003C= dev[i])\n\ncost = quicksum(fixed_cost[i] * x[i] for i in I)\npenalty = penalty_weight * quicksum(dev[i] for i in I)\nmodel.setObjective(cost + penalty, \"minimize\")\n\nmodel.setParam(\"limits\u002Ftime\", 300.0)\ntarget_gap = 0.01\nmodel.setParam(\"limits\u002Fgap\", target_gap)\nmodel.optimize()\n\nstatus = str(model.getStatus()).lower()\nif model.getNSols() == 0:\n    raise RuntimeError(f\"SCIP found no feasible solution; status={status}\")\n\nobjective = float(model.getObjVal())\ndual_bound = float(model.getDualbound())\ngap = float(model.getGap())\nquality_ok = status == \"optimal\" or gap \u003C= target_gap + 1e-12\nif not quality_ok:\n    raise RuntimeError(\n        \"SCIP has a feasible but provisional incumbent; improve it before export: \"\n        f\"status={status}, objective={objective}, dual_bound={dual_bound}, \"\n        f\"gap={gap}, target_gap={target_gap}\"\n    )\n\nselected = [i for i in I if model.getVal(x[i]) > 0.5]\n```\n\n## Common Patterns\n\n### Binary Activation\n\nUse a binary variable to allow a quantity only when an option is active.\n\n```python\nuse = {i: model.addVar(vtype=\"B\", name=f\"use_{i}\") for i in I}\nq = {i: model.addVar(lb=0, ub=upper[i], name=f\"q_{i}\") for i in I}\n\nfor i in I:\n    model.addCons(q[i] \u003C= upper[i] * use[i])\n```\n\n### Assignment\n\n```python\nassign = {\n    (i, j): model.addVar(vtype=\"B\", name=f\"assign_{i}_{j}\")\n    for i in items\n    for j in options\n}\n\nfor i in items:\n    model.addCons(quicksum(assign[i, j] for j in options) == 1)\n\nfor j in options:\n    model.addCons(quicksum(weight[i] * assign[i, j] for i in items) \u003C= capacity[j])\n```\n\n### Absolute Deviation Penalty\n\n```python\ndev = {i: model.addVar(lb=0, name=f\"dev_{i}\") for i in I}\n\nfor i in I:\n    model.addCons(actual[i] - target[i] \u003C= dev[i])\n    model.addCons(target[i] - actual[i] \u003C= dev[i])\n\npenalty_cost = penalty_weight * quicksum(dev[i] for i in I)\n```\n\n### Route Arcs\n\n```python\nSTART = \"depot_start\"\nEND = \"depot_end\"\nnodes_from = [START, *locations]\nnodes_to = [*locations, END]\narcs = [\n    (i, j)\n    for i in nodes_from\n    for j in nodes_to\n    if i != j and not (i == START and j == END)\n]\n\nx = {\n    (k, i, j): model.addVar(vtype=\"B\", name=f\"x_{k}_{i}_{j}\")\n    for k in vehicles\n    for i, j in arcs\n}\n\nfor k in vehicles:\n    model.addCons(quicksum(x[k, START, j] for j in locations) == 1)\n    model.addCons(quicksum(x[k, i, END] for i in locations) == 1)\n\n    for i in locations:\n        incoming = quicksum(x[k, j, i] for j in nodes_from if (j, i) in arcs)\n        outgoing = quicksum(x[k, i, j] for j in nodes_to if (i, j) in arcs)\n        model.addCons(incoming == outgoing)\n        model.addCons(outgoing \u003C= 1)\n```\n\nDegree and continuity constraints alone can permit disconnected cycles. Add subtour elimination for routing models.\n\n### MTZ Subtour Elimination\n\n```python\norder = {\n    (k, i): model.addVar(lb=1, ub=max(1, len(locations)), name=f\"order_{k}_{i}\")\n    for k in vehicles\n    for i in locations\n}\n\nn = len(locations)\nfor k in vehicles:\n    for i in locations:\n        for j in locations:\n            if i != j:\n                model.addCons(order[k, i] - order[k, j] + n * x[k, i, j] \u003C= n - 1)\n```\n\n## Reproducibility\n\nFix SCIP randomization and thread settings when repeatability matters.\n\n```python\ndef set_if_available(model, name, value):\n    try:\n        model.setParam(name, value)\n    except Exception:\n        pass\n\nfor name in [\n    \"randomization\u002Frandomseedshift\",\n    \"randomization\u002Fpermutationseed\",\n    \"randomization\u002Flpseed\",\n]:\n    set_if_available(model, name, 0)\n\nfor name in [\"randomization\u002Fpermutevars\", \"randomization\u002Fpermuteconss\"]:\n    set_if_available(model, name, False)\n\nset_if_available(model, \"parallel\u002Fmaxnthreads\", 1)\n```\n\n## Incumbent Quality Gate\n\nWhen the task scores objective quality, a feasible incumbent is not automatically a publishable answer. `model.getNSols() > 0` establishes only that SCIP found a feasible solution. A configured `limits\u002Fgap` is a stopping target, not evidence that the achieved gap is acceptable.\n\nBefore solving, derive and predeclare a checkable quality criterion from the public task contract: proven optimality, an explicit objective threshold, or a stated relative or absolute gap. Do not relax this gate after seeing the incumbent. If the task provides no tolerance, use a conservative small gap and the documented solve budget rather than accepting the first feasible incumbent.\n\nAfter every solve, record and inspect the status, incumbent objective, dual bound, and achieved relative gap, as shown in the template above.\n\nTreat the quality gate as failed when SCIP stops at a time or resource limit and the achieved gap or objective still misses the task criterion. In that case, keep the incumbent and continue with a task-justified larger time budget, a warm start, or a stronger formulation or cut strategy. Do not replace a documented reference budget with a substantially shorter limit merely to finish sooner when objective quality is scored.\n\nOnly serialize the final answer after the quality gate passes. The run log should retain the solver status, incumbent objective, dual bound, achieved gap, and acceptance criterion so that solution quality can be audited separately from feasibility and report arithmetic.\n\n## Extraction And Validation\n\nAfter solving, reconstruct the answer from variable values and validate it outside SCIP.\n\n```python\ndef is_selected(var):\n    return model.getVal(var) > 0.5\n\nselected_arcs = [(i, j) for i, j in arcs if is_selected(x[vehicle, i, j])]\nreported_cost = sum(distance[i, j] for i, j in selected_arcs)\n\nif abs(reported_cost - expected_cost) > 1e-6:\n    raise AssertionError(\"reported objective component does not match reconstruction\")\n```\n\nTreat SCIP feasibility as necessary but not sufficient. The final reported file still needs both the solver-quality certificate above and independent checks for schema, route reconstruction, capacity, inventory, penalties, and objective arithmetic.\n","# SCIP Optimization\n\nUse SCIP through `pyscipopt` when a task asks you to minimize or maximize an objective subject to constraints.\n\nSCIP is a strong open-source optimization solver with a Python API. It is well suited for mixed-integer optimization, routing-style models, assignment models, capacity planning, inventory movement, scheduling, and problems with soft penalties. For benchmark tasks, a SCIP-backed model plus an independent validator is usually safer than a greedy construction.\n\n## When To Use\n\nConsider PySCIPOpt when the request includes:\n\n- an objective such as minimizing cost, distance, time, unmet demand, or penalty;\n- yes\u002Fno choices, route arcs, assignments, selected items, or ordering decisions;\n- integer or continuous quantities such as load, inventory, flow, served units, or slack;\n- hard rules that every valid answer must satisfy;\n- soft rules that can be violated with an explicit penalty.\n\nDo not start by installing another optimization package. First check whether PySCIPOpt is already available:\n\n```python\ntry:\n    from pyscipopt import Model, quicksum\nexcept ImportError as exc:\n    raise RuntimeError(\"PySCIPOpt is required for this optimization approach\") from exc\n```\n\n## Modeling Workflow\n\n1. Identify sets and indices.\n   - Examples: vehicles `K`, stations `N`, jobs `J`, periods `T`, arcs `A`.\n   - Build explicit mappings when input IDs are not contiguous.\n\n2. Define decision variables.\n   - Binary variables for choices, visits, assignments, route arcs, or modes.\n   - Integer variables for counts, loads, inventory moves, or unmet units.\n   - Continuous variables for flows, costs, times, slacks, or resource levels.\n\n3. Add hard constraints.\n   - Conservation, capacity, bounds, linking, continuity, inventory limits, and mutual exclusion.\n\n4. Add soft constraints with explicit slack variables.\n   - Never use Python `abs()` on solver expressions.\n   - Linearize absolute deviation with two inequalities.\n\n5. Set a single objective.\n   - Keep named objective components such as travel cost and penalty cost.\n\n6. Solve with time and gap limits, then apply the incumbent-quality gate below.\n   - Require at least one incumbent to establish feasibility.\n   - Do not extract a final answer until the task's objective-quality criterion is met.\n\n7. Reconstruct and independently validate the output.\n   - Recompute objective components and every hard rule from the reported answer.\n\n## Minimal PySCIPOpt Template\n\n```python\nfrom pyscipopt import Model, quicksum\n\nmodel = Model(\"optimization_model\")\nmodel.hideOutput()\n\nI = range(n_items)\n\nx = {i: model.addVar(vtype=\"B\", name=f\"x_{i}\") for i in I}\namount = {\n    i: model.addVar(vtype=\"I\", lb=0, ub=capacity[i], name=f\"amount_{i}\")\n    for i in I\n}\ndev = {i: model.addVar(lb=0, name=f\"dev_{i}\") for i in I}\n\nfor i in I:\n    model.addCons(amount[i] \u003C= capacity[i] * x[i])\n    model.addCons(amount[i] - target[i] \u003C= dev[i])\n    model.addCons(target[i] - amount[i] \u003C= dev[i])\n\ncost = quicksum(fixed_cost[i] * x[i] for i in I)\npenalty = penalty_weight * quicksum(dev[i] for i in I)\nmodel.setObjective(cost + penalty, \"minimize\")\n\nmodel.setParam(\"limits\u002Ftime\", 300.0)\ntarget_gap = 0.01\nmodel.setParam(\"limits\u002Fgap\", target_gap)\nmodel.optimize()\n\nstatus = str(model.getStatus()).lower()\nif model.getNSols() == 0:\n    raise RuntimeError(f\"SCIP found no feasible solution; status={status}\")\n\nobjective = float(model.getObjVal())\ndual_bound = float(model.getDualbound())\ngap = float(model.getGap())\nquality_ok = status == \"optimal\" or gap \u003C= target_gap + 1e-12\nif not quality_ok:\n    raise RuntimeError(\n        \"SCIP has a feasible but provisional incumbent; improve it before export: \"\n        f\"status={status}, objective={objective}, dual_bound={dual_bound}, \"\n        f\"gap={gap}, target_gap={target_gap}\"\n    )\n\nselected = [i for i in I if model.getVal(x[i]) > 0.5]\n```\n\n## Common Patterns\n\n### Binary Activation\n\nUse a binary variable to allow a quantity only when an option is active.\n\n```python\nuse = {i: model.addVar(vtype=\"B\", name=f\"use_{i}\") for i in I}\nq = {i: model.addVar(lb=0, ub=upper[i], name=f\"q_{i}\") for i in I}\n\nfor i in I:\n    model.addCons(q[i] \u003C= upper[i] * use[i])\n```\n\n### Assignment\n\n```python\nassign = {\n    (i, j): model.addVar(vtype=\"B\", name=f\"assign_{i}_{j}\")\n    for i in items\n    for j in options\n}\n\nfor i in items:\n    model.addCons(quicksum(assign[i, j] for j in options) == 1)\n\nfor j in options:\n    model.addCons(quicksum(weight[i] * assign[i, j] for i in items) \u003C= capacity[j])\n```\n\n### Absolute Deviation Penalty\n\n```python\ndev = {i: model.addVar(lb=0, name=f\"dev_{i}\") for i in I}\n\nfor i in I:\n    model.addCons(actual[i] - target[i] \u003C= dev[i])\n    model.addCons(target[i] - actual[i] \u003C= dev[i])\n\npenalty_cost = penalty_weight * quicksum(dev[i] for i in I)\n```\n\n### Route Arcs\n\n```python\nSTART = \"depot_start\"\nEND = \"depot_end\"\nnodes_from = [START, *locations]\nnodes_to = [*locations, END]\narcs = [\n    (i, j)\n    for i in nodes_from\n    for j in nodes_to\n    if i != j and not (i == START and j == END)\n]\n\nx = {\n    (k, i, j): model.addVar(vtype=\"B\", name=f\"x_{k}_{i}_{j}\")\n    for k in vehicles\n    for i, j in arcs\n}\n\nfor k in vehicles:\n    model.addCons(quicksum(x[k, START, j] for j in locations) == 1)\n    model.addCons(quicksum(x[k, i, END] for i in locations) == 1)\n\n    for i in locations:\n        incoming = quicksum(x[k, j, i] for j in nodes_from if (j, i) in arcs)\n        outgoing = quicksum(x[k, i, j] for j in nodes_to if (i, j) in arcs)\n        model.addCons(incoming == outgoing)\n        model.addCons(outgoing \u003C= 1)\n```\n\nDegree and continuity constraints alone can permit disconnected cycles. Add subtour elimination for routing models.\n\n### MTZ Subtour Elimination\n\n```python\norder = {\n    (k, i): model.addVar(lb=1, ub=max(1, len(locations)), name=f\"order_{k}_{i}\")\n    for k in vehicles\n    for i in locations\n}\n\nn = len(locations)\nfor k in vehicles:\n    for i in locations:\n        for j in locations:\n            if i != j:\n                model.addCons(order[k, i] - order[k, j] + n * x[k, i, j] \u003C= n - 1)\n```\n\n## Reproducibility\n\nFix SCIP randomization and thread settings when repeatability matters.\n\n```python\ndef set_if_available(model, name, value):\n    try:\n        model.setParam(name, value)\n    except Exception:\n        pass\n\nfor name in [\n    \"randomization\u002Frandomseedshift\",\n    \"randomization\u002Fpermutationseed\",\n    \"randomization\u002Flpseed\",\n]:\n    set_if_available(model, name, 0)\n\nfor name in [\"randomization\u002Fpermutevars\", \"randomization\u002Fpermuteconss\"]:\n    set_if_available(model, name, False)\n\nset_if_available(model, \"parallel\u002Fmaxnthreads\", 1)\n```\n\n## Incumbent Quality Gate\n\nWhen the task scores objective quality, a feasible incumbent is not automatically a publishable answer. `model.getNSols() > 0` establishes only that SCIP found a feasible solution. A configured `limits\u002Fgap` is a stopping target, not evidence that the achieved gap is acceptable.\n\nBefore solving, derive and predeclare a checkable quality criterion from the public task contract: proven optimality, an explicit objective threshold, or a stated relative or absolute gap. Do not relax this gate after seeing the incumbent. If the task provides no tolerance, use a conservative small gap and the documented solve budget rather than accepting the first feasible incumbent.\n\nAfter every solve, record and inspect the status, incumbent objective, dual bound, and achieved relative gap, as shown in the template above.\n\nTreat the quality gate as failed when SCIP stops at a time or resource limit and the achieved gap or objective still misses the task criterion. In that case, keep the incumbent and continue with a task-justified larger time budget, a warm start, or a stronger formulation or cut strategy. Do not replace a documented reference budget with a substantially shorter limit merely to finish sooner when objective quality is scored.\n\nOnly serialize the final answer after the quality gate passes. The run log should retain the solver status, incumbent objective, dual bound, achieved gap, and acceptance criterion so that solution quality can be audited separately from feasibility and report arithmetic.\n\n## Extraction And Validation\n\nAfter solving, reconstruct the answer from variable values and validate it outside SCIP.\n\n```python\ndef is_selected(var):\n    return model.getVal(var) > 0.5\n\nselected_arcs = [(i, j) for i, j in arcs if is_selected(x[vehicle, i, j])]\nreported_cost = sum(distance[i, j] for i, j in selected_arcs)\n\nif abs(reported_cost - expected_cost) > 1e-6:\n    raise AssertionError(\"reported objective component does not match reconstruction\")\n```\n\nTreat SCIP feasibility as necessary but not sufficient. The final reported file still needs both the solver-quality certificate above and independent checks for schema, route reconstruction, capacity, inventory, penalties, and objective arithmetic.\n",{"name":5,"description":7},"main","cdcb6d230f3b15c3c33f8a2bdd16439f43602cbe","evaluation\u002Fdata\u002Fskillsbench87\u002Freferences\u002Fbike-rebalance\u002Fskills\u002Fscip-opt",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 scip-opt",{"tool":37,"command":38},"bunx","bunx skills add linyouxue\u002Fskill-repair-benchmark --skill scip-opt",{"tool":40,"command":41},"pnpm","pnpm dlx skills add linyouxue\u002Fskill-repair-benchmark --skill scip-opt","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Flinyouxue\u002Fskill-repair-benchmark\u002Fblob\u002Fmain\u002Fevaluation\u002Fdata\u002Fskillsbench87\u002Freferences\u002Fbike-rebalance\u002Fskills\u002Fscip-opt\u002FSKILL.md",true,[],[46,53,60,67,72,77],{"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":62,"nameZh":6,"description":63,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":64,"categories":65,"keywords":66,"publishedAt":19},"01M7E1H7DADKB6QBWKC2KQ63SZ","image-ocr","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":68,"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":69,"categories":70,"keywords":71,"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":73,"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":74,"categories":75,"keywords":76,"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":78,"name":62,"nameZh":6,"description":63,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":79,"categories":80,"keywords":81,"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",[],[]]