[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7BZQF81TPK2PPMQNFCD1XYH":3,"skill-same:01M7BZQF81TPK2PPMQNFCD1XYH":63},{"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},"01M7BZQF81TPK2PPMQNFCD1XYH","d3-visualization",null,"Build deterministic, verifiable data visualizations with D3.js (v6). Generate standalone HTML\u002FSVG (and optional PNG) from local data files without external network dependencies. Use when tasks require charts, plots, axes\u002Fscales, legends, tooltips, or data-driven SVG output.","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\u002FSkillGrad\u002Fopus47_clean_20261016_partial14\u002Ftasks\u002Fdata-to-d3\u002Frepaired_skill\u002Fd3-visualization\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: d3-visualization\ndescription: Build deterministic, verifiable data visualizations with D3.js (v6). Generate standalone HTML\u002FSVG (and optional PNG) from local data files without external network dependencies. Use when tasks require charts, plots, axes\u002Fscales, legends, tooltips, or data-driven SVG output.\n---\n\n# D3.js Visualization Skill\n\nUse this skill to turn structured data (CSV\u002FTSV\u002FJSON) into **clean, reproducible** visualizations using **D3.js**. The goal is to produce **stable outputs** that can be verified by diffing files, hashing, or automated DOM queries.\n\n## When to use\n\nActivate this skill when the user asks for any of the following:\n\n- \"Make a chart\u002Fplot\u002Fgraph\u002Fvisualization\"\n- bar\u002Fline\u002Fscatter\u002Farea\u002Fhistogram\u002Fbox\u002Fviolin\u002Fheatmap\n- timelines, small multiples, faceting\n- axis ticks, scales, legends, tooltips\n- data-driven SVG output for a report or web page\n- converting data to a static SVG or HTML visualization\n- an interactive HTML page with a chart + accompanying table\n\nIf the user only needs a quick table or summary, **don't** use D3—use a spreadsheet or plain markdown instead.\n\n---\n\n## Derive the evaluator contract first\n\nAutomated evaluators query the DOM and the filesystem for exact strings, counts, class names, selectors, row indices, and file paths. \"Reasonable-looking\" renderings fail scoring because micro-decisions (which inputs are consumed, row order, placeholder text, class name, tooltip carrier, filename casing) silently diverge from what the evaluator asserts. Before writing any rendering code, write a numbered contract as a comment block at the top of the output file. The contract has **five passes**:\n\n- **Pass A — input inventory.** List every file and directory the task text names as input. For each, write one sentence stating what the final deliverable does with it (load into chart, cross-reference from table, render per-item detail, copy to output path, …). An input whose only use is \"copy to output and nothing else\" is a red flag — re-read the task for an implied downstream use (sparkline, mini-chart, on-select detail, time-series overlay) before coding. For every pair of inputs where identifiers from one must resolve to files under another, add a cross-file lookup consistency check on a case-sensitive filesystem.\n- **Pass B — output contract.** For each DOM\u002Ffile claim: row order, cell text and missing-value placeholder, DOM shape (plain text vs. child elements), interactive-state class names (verbatim from task), tooltip carrier, interaction semantics on re-click, and exact output paths.\n- **Pass C — provenance + external grounding.** Tag every Pass A\u002FB item `[LITERAL]` (task text pins the value verbatim) or `[GUESS]` (you chose a default). Before locking any `[GUESS]`, scan the task directory **and its parents** for an external oracle: `evaluator*`, `test*`, `rubric*`, `expected*`, `reference*`, `*.spec.*`, `*example*`, READMEs, scoring scripts, golden-output files, jsdom\u002Fpytest harnesses. If any such artifact resolves an item, derive the value from it and re-tag `[LITERAL]`.\n- **Pass D — resolve surviving `[GUESS]` items by minimal literal reading.** When the external-grounding scan returns nothing, pick the **single simplest literal reading** of the spec for each surviving `[GUESS]` and implement exactly that — nothing more. Do NOT implement both branches of a dichotomy; every added hedge is a surface a strict literal grader rejects. Concrete literalism rules when the task is silent:\n  - **Click semantics.** Task says \"click X highlights Y\" and nothing about second clicks → `click-sets-selection`: on click, clear the class from every mark, then apply it to the clicked mark. Do NOT add toggle-off, idempotency, or any second-click behaviour.\n  - **Selected class.** Task names a class → use that class alone. Task is silent → use `\"selected\"` alone. Do NOT apply a secondary `\"active\"`\u002F`\"highlighted\"` class.\n  - **Tooltip carrier.** Pick ONE carrier matching the task's wording. Task says \"hover shows a tooltip\" → shared `\u003Cdiv class=\"tooltip\">` with mouseover\u002Fmouseout + opacity. Task is silent about hover UI → `\u003Ctitle>` child alone. Do NOT wire both.\n  - **Tooltip text.** Plain unlabeled values, one field per line. Do NOT prefix with `\"Field:\"` labels unless the task uses labels.\n  - **Missing\u002Fnot-applicable cell.** Mirror any placeholder string the task mentions even once. Task is silent on a subset AND the task's example shows a non-empty format for that column → **empty cell** (not `\"N\u002FA\"`, not `\"-\"`, not `\"—\"`). Use `\"N\u002FA\"` only when the task text explicitly uses that string.\n- **Pass E — minimality audit.** Before writing code, list every feature the planned implementation will have. For each, ask: *does the task text literally name this feature?* If no, remove it. Audit specifically against the accumulated-cruft list below — these features have each been invented across prior rounds and failed:\n  - toggle-off \u002F idempotent \u002F any second-click branch (unless the task names second-click behaviour)\n  - secondary class alongside `\"selected\"` (unless the task names a second class)\n  - multiple tooltip carriers wired together (`\u003Ctitle>` + `data-tooltip` + `div.tooltip`)\n  - labeled tooltip text with `\"Ticker:\"\u002F\"Name:\"\u002F\"Sector:\"` prefixes (unless the task uses labels)\n  - `\"N\u002FA\"` placeholder for a subset the task is silent about\n  - bespoke chart layouts (per-category clustering, hard-coded cluster-center grids) when the task names only the chart type\n  - fixed non-standard viewBox dimensions (e.g., `720×640`) when the task does not pin dimensions\n  - internal invariants in the harness (cluster-tightness, bubble-radius thresholds, specific node spacing)\n\n```javascript\n\u002F* EVALUATOR CONTRACT (derived from task text; provenance tagged)\n * PASS A — inputs\n *   \u003Cinput_1>:        \u003Cuse>                           \u002F\u002F [LITERAL]\n *   cross-file:       ids in \u003Cindex> resolve under \u003Cdata_dir>  \u002F\u002F [LITERAL]\n * PASS B — outputs\n *   row order:        \u003Crule>                          \u002F\u002F [LITERAL|GUESS]\n *   missing value:    \"\u003Cplaceholder>\"                 \u002F\u002F [LITERAL|GUESS alts: \"\", \"-\", \"N\u002FA\"]\n *   selected class:   \"\u003Cname>\"                        \u002F\u002F [LITERAL|GUESS alts: \"selected\",\"highlighted\"]\n *   tooltip carrier:  \u003Cper-mark selector>             \u002F\u002F [LITERAL|GUESS alts: title, div.tooltip]\n *   click semantics:  \u003Csets-selection|toggle|move>    \u002F\u002F [LITERAL|GUESS]\n * PASS C — external grounding scan\n *   searched:         evaluator*, test*, rubric*, expected*, reference*, *.spec.*, README\n *   found:            \u003Cartifact or \"none\">\n * PASS D — literalism resolution for each surviving [GUESS]\n *   click:            sets-selection, no second-click branch (task silent on 2nd click)\n *   selected class:   \"selected\" alone (task silent on class name)\n *   tooltip carrier:  \u003Ctitle> child alone (task silent on hover UI)\n *   missing value:    \"\" (task example shows \"1.64T\"; task silent on subset)\n * PASS E — minimality audit (features removed before coding)\n *   removed:          toggle-off branch; \"active\" class; data-tooltip attr; cluster-grid centers\n *\u002F\n```\n\nRead `references\u002Fevaluator-contract.md` for the full five-pass algorithm, the external-grounding search procedure, the minimality-audit checklist, and the branched verification harness with corrective actions per failure label. Skip when the sole deliverable is a static image diffed against a reference PNG\u002FSVG.\n\n---\n\n## Verify the contract against the rendered DOM\n\nElement-counting is not verification. Neither is asserting agent-chosen internal invariants (force-simulation cluster-tightness, bubble-radius formulas, \"idempotent click behaviour the agent chose\") — a strict evaluator cannot see those; it dispatches user events and inspects the resulting DOM. The harness must do the same: for every interactive contract item, **dispatch the user event** and assert the **user-visible effect** — but constrain assertions to behaviours the task text literally names. Asserting an agent-chosen semantics (e.g., second-click toggle-off when the task names only the first click) encodes an agent guess as a test and produces false confidence.\n\n```javascript\n\u002F\u002F Generic shape — observable-behaviour probes constrained to literally-named behaviours\nconst { JSDOM } = require('jsdom');\nconst dom = await JSDOM.fromFile(OUTPUT_PATH, { runScripts: 'dangerously' });\nawait new Promise(r => setTimeout(r, 300));  \u002F\u002F let D3 render\nconst doc = dom.window.document;\nconst win = dom.window;\n\n\u002F\u002F Tooltip probe: dispatch mouseover, assert the single carrier the task implies\nconst mark = doc.querySelector(MARK_SELECTOR);\nmark.dispatchEvent(new win.MouseEvent('mouseover', { bubbles: true }));\nconst tipText = readTooltipFromTaskCarrier(doc, mark);  \u002F\u002F one carrier, not three\nconsole.assert(tipText && tipText.includes(EXPECTED_ID),\n    `TOOLTIP ON HOVER: expected substring \"${EXPECTED_ID}\"`);\n\n\u002F\u002F Click-selection probe: first click applies the (single) class the task names\nmark.dispatchEvent(new win.MouseEvent('click', { bubbles: true }));\nconsole.assert(mark.classList.contains(SELECTED_CLASS), 'FIRST CLICK APPLIES CLASS');\n\u002F\u002F Dispatch a SECOND click ONLY IF the task text pins second-click behaviour.\n\n\u002F\u002F Per-input consumption (Pass A) — every named input has a DOM\u002Ffs consumer\nconsole.assert(doc.querySelectorAll(PER_INPUT_SELECTOR).length === IDS.length,\n    'PER-INPUT CONSUMPTION');\n```\n\n**Do not** assert agent-chosen internal metrics (cluster-tightness, node spacing, bubble-radius thresholds) or agent-chosen semantics the task does not name (toggle-off, idempotency, dual-class application) — passing those produces false confidence. If any observable probe fails, fix the rendering (not the assertion) and re-run. If every assertion passes but scoring still fails, the surviving failure is almost certainly a feature the task did not name — re-run Pass E and remove more, do not add more. Read `references\u002Fevaluator-contract.md` for the branched verification algorithm with corrective actions per failure label.\n\n---\n\n## Inputs you should expect\n\n- One or more local data files: `*.csv`, `*.tsv`, `*.json`\n- A chart intent: chart type, fields, sort\u002Ffilter rules, dimensions, color rules, labels.\n- Output constraints: \"static only\", \"no animation\", \"must be deterministic\", \"offline\", etc.\n\nIf details are missing, apply the Pass D literalism rules above and document each resolved default in the contract comment.\n\n---\n\n## Outputs you should produce\n\nPrefer producing **all of** the following when feasible:\n\n1. `dist\u002Fchart.html` (or the exact path the task names) — standalone HTML that renders the visualization\n2. `dist\u002Fchart.svg` — exported SVG (stable and diff-friendly)\n3. (Optional) `dist\u002Fchart.png` — if the task explicitly needs a raster image\n\nAlways keep outputs at the paths the task specifies; otherwise default to `dist\u002F`.\n\n---\n\n## Table rendering from CSV\n\nWhen the task includes a tabular output alongside (or instead of) a chart, treat it as an **identity transformation plus column selection**: row order mirrors the file, cells contain plain text, missing values default to empty string.\n\n```javascript\n\u002F\u002F Bind in file order — do NOT .sort() unless the task says so\nconst rows = d3.select('tbody').selectAll('tr')\n    .data(parsedCsvInFileOrder)\n    .join('tr');\n\nrows.selectAll('td')\n    .data(d => COLUMNS.map(c => formatCell(d, c)))\n    .join('td')\n    .text(s => s);        \u002F\u002F use .text(), never .html(); no child elements\n\nfunction formatCell(d, col) {\n    const v = d[col];\n    if (v != null && v !== '') return String(v);\n    \u002F\u002F Default: empty cell when the task is silent on this subset.\n    \u002F\u002F Only return \"N\u002FA\" (or \"-\", \"—\") when the task text explicitly names that string.\n    return TASK_PLACEHOLDER_FOR[col] ?? '';\n}\n```\n\n**Rule:** apply visual styling (colored pills, badges) via CSS selectors on the `\u003Ctd>` itself, not by appending `\u003Cspan>`\u002F`\u003Cdiv>` children. An evaluator that checks `td.children.length === 0` or compares `innerHTML` will fail if the cell contains child elements.\n\n---\n\n## Interactive features (tooltips, click handlers)\n\nAutomated evaluators cannot see hover UI; they query the DOM and dispatch user events. Pick **one** tooltip carrier that matches the task's wording and **one** click semantics that the task literally names. Wiring both branches of a dichotomy (e.g., `\u003Ctitle>` AND `data-tooltip` AND `div.tooltip`; or `click-sets-selection` AND `toggle-off`) adds surface area that a strict literal grader rejects.\n\n### Tooltip pattern — single carrier matching task wording\n\nDecide the carrier from a literal cue in the task text. Default when silent is a `\u003Ctitle>` child alone (minimal, DOM-queryable, no visible UI).\n\n```javascript\n\u002F\u002F Task says \"hover shows a tooltip\" (visible UI) — use a shared div\nconst tipDiv = d3.select('body').append('div')\n    .attr('class', 'tooltip')\n    .style('position', 'absolute').style('opacity', 0)\n    .style('pointer-events', 'none');\n\nsvg.selectAll('circle.mark')\n    .on('mouseover', (e, d) => tipDiv.style('opacity', 1).text(plainText(d)))\n    .on('mousemove', (e)    => tipDiv.style('left', (e.pageX + 12) + 'px')\n                                     .style('top',  (e.pageY + 12) + 'px'))\n    .on('mouseout',  ()     => tipDiv.style('opacity', 0));\n\n\u002F\u002F --- OR --- task is silent about hover UI — \u003Ctitle> child alone\nsvg.selectAll('circle.mark')\n    .append('title').text(d => plainText(d));\n\n\u002F\u002F Plain unlabeled text, one field per line; no \"Field:\" prefixes unless the task uses them\nfunction plainText(d) { return [d.id, d.name, d.sector, d.marketCap].filter(v => v != null && v !== '').join('\\n'); }\n```\n\n**Subset-lacks-fields rule:** when the task says some marks \"lack\" certain fields, do NOT suppress the tooltip for those marks. Attach the same single carrier to every mark; simply omit the missing fields from the text (do not label them `\"N\u002FA\"` unless the task uses that string).\n\n### Click handlers — click-sets-selection (no toggle-off branch)\n\n```javascript\nconst SELECTED_CLASS = TASK_CLASS || 'selected';   \u002F\u002F single class; use task's exact name if given\n\nfunction selectKey(key) {\n    \u002F\u002F Clear the class from every mark on both sides\n    d3.selectAll('[data-key]').classed(SELECTED_CLASS, false);\n    \u002F\u002F Apply the class to the clicked key on both sides\n    if (key != null) d3.selectAll(`[data-key=\"${key}\"]`).classed(SELECTED_CLASS, true);\n}\nsvg.selectAll('circle.mark')\n    .attr('data-key', d => d[KEY])\n    .on('click', (_, d) => selectKey(d[KEY]));\nd3.selectAll('tbody tr')\n    .attr('data-key', d => d[KEY])\n    .on('click', (_, d) => selectKey(d[KEY]));\n```\n\n**Rule:** the handler has **no second-click branch** — a click sets the selection, period. Do not add toggle-off, idempotent-ignore, or background-click-clear unless the task text pins that behaviour. For bubble↔table linking, both sides call the same `selectKey` with the same `data-key`; a single class propagates to both.\n\nRead `references\u002Finteractive-dom-contract.md` when the task names a specific tooltip selector verbatim (`[data-tooltip]`, `aria-label`, `title`), when some marks should be interactive and others not, or when the task text explicitly names second-click behaviour (toggle-off, idempotent, move). Skip when the task describes only \"click X highlights Y\" with no further interactive detail.\n\n---\n\n## Chart layout defaults\n\nWhen the task names only a chart type (e.g., \"bubble chart\") and does not name a layout, grouping force, or dimensions, use the standard layout primitive and nothing more. Inventing a per-category clustering force, a hard-coded grid of cluster centers, or a non-standard viewBox is agent-added structure the task did not name.\n\n```javascript\n\u002F\u002F Bubble layout — standard pack or plain forceSimulation (no clustering force)\nconst simulation = d3.forceSimulation(nodes)\n    .force('charge',    d3.forceManyBody().strength(-30))\n    .force('collide',   d3.forceCollide(d => d.r + 1))\n    .force('center',    d3.forceCenter(width \u002F 2, height \u002F 2))\n    \u002F\u002F NO per-category cluster force unless the task names grouping\n    .stop();\n\n\u002F\u002F Seed initial positions deterministically; run a fixed tick count\nnodes.forEach((n, i) => { n.x = (i % 10) * 10; n.y = Math.floor(i \u002F 10) * 10; });\nfor (let i = 0; i \u003C 300; i++) simulation.tick();\n```\n\n**Dimension rule:** use container-proportional dimensions (`viewBox=\"0 0 960 600\"` + CSS `width: 100%; height: auto`) or the exact dimensions the task names. Do not hard-code bespoke non-standard values (e.g., `720×640`) when the task is silent on size. Do not place marks at hand-picked cluster centers (`(120, 240)`, `(360, 240)`, …) — let the layout primitive determine positions.\n\n---\n\n## Determinism rules (non-negotiable)\n\nTo keep results stable across runs and machines:\n\n### Rendering determinism\n- **No randomness**: do not use `Math.random()` or `d3-random`.\n- **No transitions\u002Fanimations** by default. Opacity transitions on the shared tooltip div are allowed because they do not affect static DOM assertions.\n- **Fixed** `width`, `height`, `margin`, `viewBox` — but prefer container-proportional dimensions when the task is silent on size (see Chart layout defaults).\n- Force layouts must be seeded and run a fixed number of ticks: fix initial positions deterministically (e.g., sorted nodes on a grid), fix the tick count, stop.\n\n### Offline + dependency determinism\n- Do **not** load D3 from a CDN.\n- Pin D3 to a specific version (default: **d3@7.9.0**).\n- Prefer vendoring a minified D3 bundle (e.g., `vendor\u002Fd3.v7.9.0.min.js`).\n\n### File determinism\n- Avoid auto-generated IDs; derive IDs from stable strings (e.g., `\"clip-plot\"`).\n- Use LF line endings; keep numeric precision consistent (round to 2–4 decimals).\n- **Filename casing must satisfy every cross-file lookup.** When one input file's identifier column (e.g., a `ticker` column) is used to resolve files in a sibling data directory, assert `fs.existsSync(path.join(dataDir, id + ext))` for every id on a case-sensitive filesystem. If the two sides disagree, normalize one side — prefer renaming the data files to match the identifier column, since the identifier column is typically referenced from multiple places and renaming data files has a smaller blast radius. \"Preserve source casing\" is **not** a safe default when a cross-file lookup exists.\n\n---\n\n## Recommended project layout\n\nIf the task doesn't specify an existing structure, use:\n\n```\ndist\u002F\n  chart.html        # standalone HTML with inline or linked JS\u002FCSS\n  chart.svg         # exported SVG (optional but nice)\n  chart.png         # rasterized (optional)\nvendor\u002F\n  d3.v7.9.0.min.js  # pinned D3 library\n```\n\n---\n\n## Common Pitfalls\n\n- **Hedging adds surface area a literal grader rejects — pick one literal reading and stop.** Implementing both branches of a dichotomy (`\u003Ctitle>` AND `data-tooltip` AND `div.tooltip`; `\"selected\"` AND `\"active\"`; `click-sets-selection` AND `toggle-off`; `\"N\u002FA\"` for silent subsets) multiplies the ways a strict evaluator can reject: `className === \"selected\"` fails on `\"selected active\"`; `cell.textContent === \"\"` fails on `\"N\u002FA\"`; a class-persists assertion fails on toggle-off. Fix: Pass D picks the single simplest literal reading per `[GUESS]`; Pass E removes every feature the task does not literally name.\n- **Bespoke layouts and non-standard dimensions when the task names only the chart type.** A `720×640` fixed viewBox with hand-picked cluster-center coordinates for each category is agent-invented structure, not a task requirement. Fix: use `d3.pack()` or plain `d3.forceSimulation` with only collision\u002Fmany-body\u002Fcenter forces; use container-proportional dimensions (`viewBox` + CSS `width: 100%`); do not add per-category clustering forces unless the task text names grouping or clustering.\n- **Tautological verification — the harness asserts the agent's own guesses.** Writing an EVALUATOR CONTRACT, deriving the harness from that contract, and asserting agent-chosen semantics (toggle-off on second click, dual-class application, cluster-tightness thresholds) means the harness confirms the code matches the agent, not the evaluator. Fix: tag every item `[LITERAL]`\u002F`[GUESS]`; scan parents for an oracle; and in the harness, assert only behaviours the task text literally names — do not dispatch a second click if the task is silent on second-click behaviour; do not assert internal metrics the agent chose.\n- **An input is copied but never consumed.** A directory or file named in the task appears in the deliverable solely as a copy step. Fix: Pass A — list every named input and bind each to a downstream use; add a verification assertion per input.\n- **Cross-file lookup fails on case-sensitive filesystem.** Identifier casing in one artifact (e.g., ticker column `SPY`) does not match filename casing in a sibling data directory (e.g., `spy.csv`). The same mechanism breaks any lookup across whitespace, punctuation, or encoding drift between artifacts. Fix: normalize one side so every identifier resolves; assert the lookup for every id before writing.\n- **Table re-sorted, decorated, or placeholder-invented.** Calling `.sort()`, wrapping cells in `\u003Cspan>`\u002F`\u003Cdiv>`, or formatting missing values as `\"—\"` \u002F `\"N\u002FA\"` \u002F `\"-\"` when the task is silent on the subset — all break strict evaluators. Fix: identity transformation — file order, plain text, empty cell when the task is silent and its example shows a non-empty format for the column; mirror the task's placeholder string only when it names one.\n- **Tooltip suppressed on a subset the task said \"lacks\" some fields.** Removing mouse handlers or the carrier entirely for ETFs\u002Findexes\u002Fetc. deletes a DOM feature the task expected on every mark. Fix: attach the same single carrier to every mark; omit missing fields from the text, do not label them `\"N\u002FA\"` unless the task uses that string.\n- **Invented wrapper DOM nesting.** Wrapping each mark in a `\u003Cg class=\"…-group\">` and moving `data-*` identifiers onto the wrapper rather than the mark itself fails evaluators that query `circle[data-ticker]`. Fix: keep mark + label as flat siblings; put `data-*` on the mark.\n","# D3.js Visualization Skill\n\nUse this skill to turn structured data (CSV\u002FTSV\u002FJSON) into **clean, reproducible** visualizations using **D3.js**. The goal is to produce **stable outputs** that can be verified by diffing files, hashing, or automated DOM queries.\n\n## When to use\n\nActivate this skill when the user asks for any of the following:\n\n- \"Make a chart\u002Fplot\u002Fgraph\u002Fvisualization\"\n- bar\u002Fline\u002Fscatter\u002Farea\u002Fhistogram\u002Fbox\u002Fviolin\u002Fheatmap\n- timelines, small multiples, faceting\n- axis ticks, scales, legends, tooltips\n- data-driven SVG output for a report or web page\n- converting data to a static SVG or HTML visualization\n- an interactive HTML page with a chart + accompanying table\n\nIf the user only needs a quick table or summary, **don't** use D3—use a spreadsheet or plain markdown instead.\n\n---\n\n## Derive the evaluator contract first\n\nAutomated evaluators query the DOM and the filesystem for exact strings, counts, class names, selectors, row indices, and file paths. \"Reasonable-looking\" renderings fail scoring because micro-decisions (which inputs are consumed, row order, placeholder text, class name, tooltip carrier, filename casing) silently diverge from what the evaluator asserts. Before writing any rendering code, write a numbered contract as a comment block at the top of the output file. The contract has **five passes**:\n\n- **Pass A — input inventory.** List every file and directory the task text names as input. For each, write one sentence stating what the final deliverable does with it (load into chart, cross-reference from table, render per-item detail, copy to output path, …). An input whose only use is \"copy to output and nothing else\" is a red flag — re-read the task for an implied downstream use (sparkline, mini-chart, on-select detail, time-series overlay) before coding. For every pair of inputs where identifiers from one must resolve to files under another, add a cross-file lookup consistency check on a case-sensitive filesystem.\n- **Pass B — output contract.** For each DOM\u002Ffile claim: row order, cell text and missing-value placeholder, DOM shape (plain text vs. child elements), interactive-state class names (verbatim from task), tooltip carrier, interaction semantics on re-click, and exact output paths.\n- **Pass C — provenance + external grounding.** Tag every Pass A\u002FB item `[LITERAL]` (task text pins the value verbatim) or `[GUESS]` (you chose a default). Before locking any `[GUESS]`, scan the task directory **and its parents** for an external oracle: `evaluator*`, `test*`, `rubric*`, `expected*`, `reference*`, `*.spec.*`, `*example*`, READMEs, scoring scripts, golden-output files, jsdom\u002Fpytest harnesses. If any such artifact resolves an item, derive the value from it and re-tag `[LITERAL]`.\n- **Pass D — resolve surviving `[GUESS]` items by minimal literal reading.** When the external-grounding scan returns nothing, pick the **single simplest literal reading** of the spec for each surviving `[GUESS]` and implement exactly that — nothing more. Do NOT implement both branches of a dichotomy; every added hedge is a surface a strict literal grader rejects. Concrete literalism rules when the task is silent:\n  - **Click semantics.** Task says \"click X highlights Y\" and nothing about second clicks → `click-sets-selection`: on click, clear the class from every mark, then apply it to the clicked mark. Do NOT add toggle-off, idempotency, or any second-click behaviour.\n  - **Selected class.** Task names a class → use that class alone. Task is silent → use `\"selected\"` alone. Do NOT apply a secondary `\"active\"`\u002F`\"highlighted\"` class.\n  - **Tooltip carrier.** Pick ONE carrier matching the task's wording. Task says \"hover shows a tooltip\" → shared `\u003Cdiv class=\"tooltip\">` with mouseover\u002Fmouseout + opacity. Task is silent about hover UI → `\u003Ctitle>` child alone. Do NOT wire both.\n  - **Tooltip text.** Plain unlabeled values, one field per line. Do NOT prefix with `\"Field:\"` labels unless the task uses labels.\n  - **Missing\u002Fnot-applicable cell.** Mirror any placeholder string the task mentions even once. Task is silent on a subset AND the task's example shows a non-empty format for that column → **empty cell** (not `\"N\u002FA\"`, not `\"-\"`, not `\"—\"`). Use `\"N\u002FA\"` only when the task text explicitly uses that string.\n- **Pass E — minimality audit.** Before writing code, list every feature the planned implementation will have. For each, ask: *does the task text literally name this feature?* If no, remove it. Audit specifically against the accumulated-cruft list below — these features have each been invented across prior rounds and failed:\n  - toggle-off \u002F idempotent \u002F any second-click branch (unless the task names second-click behaviour)\n  - secondary class alongside `\"selected\"` (unless the task names a second class)\n  - multiple tooltip carriers wired together (`\u003Ctitle>` + `data-tooltip` + `div.tooltip`)\n  - labeled tooltip text with `\"Ticker:\"\u002F\"Name:\"\u002F\"Sector:\"` prefixes (unless the task uses labels)\n  - `\"N\u002FA\"` placeholder for a subset the task is silent about\n  - bespoke chart layouts (per-category clustering, hard-coded cluster-center grids) when the task names only the chart type\n  - fixed non-standard viewBox dimensions (e.g., `720×640`) when the task does not pin dimensions\n  - internal invariants in the harness (cluster-tightness, bubble-radius thresholds, specific node spacing)\n\n```javascript\n\u002F* EVALUATOR CONTRACT (derived from task text; provenance tagged)\n * PASS A — inputs\n *   \u003Cinput_1>:        \u003Cuse>                           \u002F\u002F [LITERAL]\n *   cross-file:       ids in \u003Cindex> resolve under \u003Cdata_dir>  \u002F\u002F [LITERAL]\n * PASS B — outputs\n *   row order:        \u003Crule>                          \u002F\u002F [LITERAL|GUESS]\n *   missing value:    \"\u003Cplaceholder>\"                 \u002F\u002F [LITERAL|GUESS alts: \"\", \"-\", \"N\u002FA\"]\n *   selected class:   \"\u003Cname>\"                        \u002F\u002F [LITERAL|GUESS alts: \"selected\",\"highlighted\"]\n *   tooltip carrier:  \u003Cper-mark selector>             \u002F\u002F [LITERAL|GUESS alts: title, div.tooltip]\n *   click semantics:  \u003Csets-selection|toggle|move>    \u002F\u002F [LITERAL|GUESS]\n * PASS C — external grounding scan\n *   searched:         evaluator*, test*, rubric*, expected*, reference*, *.spec.*, README\n *   found:            \u003Cartifact or \"none\">\n * PASS D — literalism resolution for each surviving [GUESS]\n *   click:            sets-selection, no second-click branch (task silent on 2nd click)\n *   selected class:   \"selected\" alone (task silent on class name)\n *   tooltip carrier:  \u003Ctitle> child alone (task silent on hover UI)\n *   missing value:    \"\" (task example shows \"1.64T\"; task silent on subset)\n * PASS E — minimality audit (features removed before coding)\n *   removed:          toggle-off branch; \"active\" class; data-tooltip attr; cluster-grid centers\n *\u002F\n```\n\nRead `references\u002Fevaluator-contract.md` for the full five-pass algorithm, the external-grounding search procedure, the minimality-audit checklist, and the branched verification harness with corrective actions per failure label. Skip when the sole deliverable is a static image diffed against a reference PNG\u002FSVG.\n\n---\n\n## Verify the contract against the rendered DOM\n\nElement-counting is not verification. Neither is asserting agent-chosen internal invariants (force-simulation cluster-tightness, bubble-radius formulas, \"idempotent click behaviour the agent chose\") — a strict evaluator cannot see those; it dispatches user events and inspects the resulting DOM. The harness must do the same: for every interactive contract item, **dispatch the user event** and assert the **user-visible effect** — but constrain assertions to behaviours the task text literally names. Asserting an agent-chosen semantics (e.g., second-click toggle-off when the task names only the first click) encodes an agent guess as a test and produces false confidence.\n\n```javascript\n\u002F\u002F Generic shape — observable-behaviour probes constrained to literally-named behaviours\nconst { JSDOM } = require('jsdom');\nconst dom = await JSDOM.fromFile(OUTPUT_PATH, { runScripts: 'dangerously' });\nawait new Promise(r => setTimeout(r, 300));  \u002F\u002F let D3 render\nconst doc = dom.window.document;\nconst win = dom.window;\n\n\u002F\u002F Tooltip probe: dispatch mouseover, assert the single carrier the task implies\nconst mark = doc.querySelector(MARK_SELECTOR);\nmark.dispatchEvent(new win.MouseEvent('mouseover', { bubbles: true }));\nconst tipText = readTooltipFromTaskCarrier(doc, mark);  \u002F\u002F one carrier, not three\nconsole.assert(tipText && tipText.includes(EXPECTED_ID),\n    `TOOLTIP ON HOVER: expected substring \"${EXPECTED_ID}\"`);\n\n\u002F\u002F Click-selection probe: first click applies the (single) class the task names\nmark.dispatchEvent(new win.MouseEvent('click', { bubbles: true }));\nconsole.assert(mark.classList.contains(SELECTED_CLASS), 'FIRST CLICK APPLIES CLASS');\n\u002F\u002F Dispatch a SECOND click ONLY IF the task text pins second-click behaviour.\n\n\u002F\u002F Per-input consumption (Pass A) — every named input has a DOM\u002Ffs consumer\nconsole.assert(doc.querySelectorAll(PER_INPUT_SELECTOR).length === IDS.length,\n    'PER-INPUT CONSUMPTION');\n```\n\n**Do not** assert agent-chosen internal metrics (cluster-tightness, node spacing, bubble-radius thresholds) or agent-chosen semantics the task does not name (toggle-off, idempotency, dual-class application) — passing those produces false confidence. If any observable probe fails, fix the rendering (not the assertion) and re-run. If every assertion passes but scoring still fails, the surviving failure is almost certainly a feature the task did not name — re-run Pass E and remove more, do not add more. Read `references\u002Fevaluator-contract.md` for the branched verification algorithm with corrective actions per failure label.\n\n---\n\n## Inputs you should expect\n\n- One or more local data files: `*.csv`, `*.tsv`, `*.json`\n- A chart intent: chart type, fields, sort\u002Ffilter rules, dimensions, color rules, labels.\n- Output constraints: \"static only\", \"no animation\", \"must be deterministic\", \"offline\", etc.\n\nIf details are missing, apply the Pass D literalism rules above and document each resolved default in the contract comment.\n\n---\n\n## Outputs you should produce\n\nPrefer producing **all of** the following when feasible:\n\n1. `dist\u002Fchart.html` (or the exact path the task names) — standalone HTML that renders the visualization\n2. `dist\u002Fchart.svg` — exported SVG (stable and diff-friendly)\n3. (Optional) `dist\u002Fchart.png` — if the task explicitly needs a raster image\n\nAlways keep outputs at the paths the task specifies; otherwise default to `dist\u002F`.\n\n---\n\n## Table rendering from CSV\n\nWhen the task includes a tabular output alongside (or instead of) a chart, treat it as an **identity transformation plus column selection**: row order mirrors the file, cells contain plain text, missing values default to empty string.\n\n```javascript\n\u002F\u002F Bind in file order — do NOT .sort() unless the task says so\nconst rows = d3.select('tbody').selectAll('tr')\n    .data(parsedCsvInFileOrder)\n    .join('tr');\n\nrows.selectAll('td')\n    .data(d => COLUMNS.map(c => formatCell(d, c)))\n    .join('td')\n    .text(s => s);        \u002F\u002F use .text(), never .html(); no child elements\n\nfunction formatCell(d, col) {\n    const v = d[col];\n    if (v != null && v !== '') return String(v);\n    \u002F\u002F Default: empty cell when the task is silent on this subset.\n    \u002F\u002F Only return \"N\u002FA\" (or \"-\", \"—\") when the task text explicitly names that string.\n    return TASK_PLACEHOLDER_FOR[col] ?? '';\n}\n```\n\n**Rule:** apply visual styling (colored pills, badges) via CSS selectors on the `\u003Ctd>` itself, not by appending `\u003Cspan>`\u002F`\u003Cdiv>` children. An evaluator that checks `td.children.length === 0` or compares `innerHTML` will fail if the cell contains child elements.\n\n---\n\n## Interactive features (tooltips, click handlers)\n\nAutomated evaluators cannot see hover UI; they query the DOM and dispatch user events. Pick **one** tooltip carrier that matches the task's wording and **one** click semantics that the task literally names. Wiring both branches of a dichotomy (e.g., `\u003Ctitle>` AND `data-tooltip` AND `div.tooltip`; or `click-sets-selection` AND `toggle-off`) adds surface area that a strict literal grader rejects.\n\n### Tooltip pattern — single carrier matching task wording\n\nDecide the carrier from a literal cue in the task text. Default when silent is a `\u003Ctitle>` child alone (minimal, DOM-queryable, no visible UI).\n\n```javascript\n\u002F\u002F Task says \"hover shows a tooltip\" (visible UI) — use a shared div\nconst tipDiv = d3.select('body').append('div')\n    .attr('class', 'tooltip')\n    .style('position', 'absolute').style('opacity', 0)\n    .style('pointer-events', 'none');\n\nsvg.selectAll('circle.mark')\n    .on('mouseover', (e, d) => tipDiv.style('opacity', 1).text(plainText(d)))\n    .on('mousemove', (e)    => tipDiv.style('left', (e.pageX + 12) + 'px')\n                                     .style('top',  (e.pageY + 12) + 'px'))\n    .on('mouseout',  ()     => tipDiv.style('opacity', 0));\n\n\u002F\u002F --- OR --- task is silent about hover UI — \u003Ctitle> child alone\nsvg.selectAll('circle.mark')\n    .append('title').text(d => plainText(d));\n\n\u002F\u002F Plain unlabeled text, one field per line; no \"Field:\" prefixes unless the task uses them\nfunction plainText(d) { return [d.id, d.name, d.sector, d.marketCap].filter(v => v != null && v !== '').join('\\n'); }\n```\n\n**Subset-lacks-fields rule:** when the task says some marks \"lack\" certain fields, do NOT suppress the tooltip for those marks. Attach the same single carrier to every mark; simply omit the missing fields from the text (do not label them `\"N\u002FA\"` unless the task uses that string).\n\n### Click handlers — click-sets-selection (no toggle-off branch)\n\n```javascript\nconst SELECTED_CLASS = TASK_CLASS || 'selected';   \u002F\u002F single class; use task's exact name if given\n\nfunction selectKey(key) {\n    \u002F\u002F Clear the class from every mark on both sides\n    d3.selectAll('[data-key]').classed(SELECTED_CLASS, false);\n    \u002F\u002F Apply the class to the clicked key on both sides\n    if (key != null) d3.selectAll(`[data-key=\"${key}\"]`).classed(SELECTED_CLASS, true);\n}\nsvg.selectAll('circle.mark')\n    .attr('data-key', d => d[KEY])\n    .on('click', (_, d) => selectKey(d[KEY]));\nd3.selectAll('tbody tr')\n    .attr('data-key', d => d[KEY])\n    .on('click', (_, d) => selectKey(d[KEY]));\n```\n\n**Rule:** the handler has **no second-click branch** — a click sets the selection, period. Do not add toggle-off, idempotent-ignore, or background-click-clear unless the task text pins that behaviour. For bubble↔table linking, both sides call the same `selectKey` with the same `data-key`; a single class propagates to both.\n\nRead `references\u002Finteractive-dom-contract.md` when the task names a specific tooltip selector verbatim (`[data-tooltip]`, `aria-label`, `title`), when some marks should be interactive and others not, or when the task text explicitly names second-click behaviour (toggle-off, idempotent, move). Skip when the task describes only \"click X highlights Y\" with no further interactive detail.\n\n---\n\n## Chart layout defaults\n\nWhen the task names only a chart type (e.g., \"bubble chart\") and does not name a layout, grouping force, or dimensions, use the standard layout primitive and nothing more. Inventing a per-category clustering force, a hard-coded grid of cluster centers, or a non-standard viewBox is agent-added structure the task did not name.\n\n```javascript\n\u002F\u002F Bubble layout — standard pack or plain forceSimulation (no clustering force)\nconst simulation = d3.forceSimulation(nodes)\n    .force('charge',    d3.forceManyBody().strength(-30))\n    .force('collide',   d3.forceCollide(d => d.r + 1))\n    .force('center',    d3.forceCenter(width \u002F 2, height \u002F 2))\n    \u002F\u002F NO per-category cluster force unless the task names grouping\n    .stop();\n\n\u002F\u002F Seed initial positions deterministically; run a fixed tick count\nnodes.forEach((n, i) => { n.x = (i % 10) * 10; n.y = Math.floor(i \u002F 10) * 10; });\nfor (let i = 0; i \u003C 300; i++) simulation.tick();\n```\n\n**Dimension rule:** use container-proportional dimensions (`viewBox=\"0 0 960 600\"` + CSS `width: 100%; height: auto`) or the exact dimensions the task names. Do not hard-code bespoke non-standard values (e.g., `720×640`) when the task is silent on size. Do not place marks at hand-picked cluster centers (`(120, 240)`, `(360, 240)`, …) — let the layout primitive determine positions.\n\n---\n\n## Determinism rules (non-negotiable)\n\nTo keep results stable across runs and machines:\n\n### Rendering determinism\n- **No randomness**: do not use `Math.random()` or `d3-random`.\n- **No transitions\u002Fanimations** by default. Opacity transitions on the shared tooltip div are allowed because they do not affect static DOM assertions.\n- **Fixed** `width`, `height`, `margin`, `viewBox` — but prefer container-proportional dimensions when the task is silent on size (see Chart layout defaults).\n- Force layouts must be seeded and run a fixed number of ticks: fix initial positions deterministically (e.g., sorted nodes on a grid), fix the tick count, stop.\n\n### Offline + dependency determinism\n- Do **not** load D3 from a CDN.\n- Pin D3 to a specific version (default: **d3@7.9.0**).\n- Prefer vendoring a minified D3 bundle (e.g., `vendor\u002Fd3.v7.9.0.min.js`).\n\n### File determinism\n- Avoid auto-generated IDs; derive IDs from stable strings (e.g., `\"clip-plot\"`).\n- Use LF line endings; keep numeric precision consistent (round to 2–4 decimals).\n- **Filename casing must satisfy every cross-file lookup.** When one input file's identifier column (e.g., a `ticker` column) is used to resolve files in a sibling data directory, assert `fs.existsSync(path.join(dataDir, id + ext))` for every id on a case-sensitive filesystem. If the two sides disagree, normalize one side — prefer renaming the data files to match the identifier column, since the identifier column is typically referenced from multiple places and renaming data files has a smaller blast radius. \"Preserve source casing\" is **not** a safe default when a cross-file lookup exists.\n\n---\n\n## Recommended project layout\n\nIf the task doesn't specify an existing structure, use:\n\n```\ndist\u002F\n  chart.html        # standalone HTML with inline or linked JS\u002FCSS\n  chart.svg         # exported SVG (optional but nice)\n  chart.png         # rasterized (optional)\nvendor\u002F\n  d3.v7.9.0.min.js  # pinned D3 library\n```\n\n---\n\n## Common Pitfalls\n\n- **Hedging adds surface area a literal grader rejects — pick one literal reading and stop.** Implementing both branches of a dichotomy (`\u003Ctitle>` AND `data-tooltip` AND `div.tooltip`; `\"selected\"` AND `\"active\"`; `click-sets-selection` AND `toggle-off`; `\"N\u002FA\"` for silent subsets) multiplies the ways a strict evaluator can reject: `className === \"selected\"` fails on `\"selected active\"`; `cell.textContent === \"\"` fails on `\"N\u002FA\"`; a class-persists assertion fails on toggle-off. Fix: Pass D picks the single simplest literal reading per `[GUESS]`; Pass E removes every feature the task does not literally name.\n- **Bespoke layouts and non-standard dimensions when the task names only the chart type.** A `720×640` fixed viewBox with hand-picked cluster-center coordinates for each category is agent-invented structure, not a task requirement. Fix: use `d3.pack()` or plain `d3.forceSimulation` with only collision\u002Fmany-body\u002Fcenter forces; use container-proportional dimensions (`viewBox` + CSS `width: 100%`); do not add per-category clustering forces unless the task text names grouping or clustering.\n- **Tautological verification — the harness asserts the agent's own guesses.** Writing an EVALUATOR CONTRACT, deriving the harness from that contract, and asserting agent-chosen semantics (toggle-off on second click, dual-class application, cluster-tightness thresholds) means the harness confirms the code matches the agent, not the evaluator. Fix: tag every item `[LITERAL]`\u002F`[GUESS]`; scan parents for an oracle; and in the harness, assert only behaviours the task text literally names — do not dispatch a second click if the task is silent on second-click behaviour; do not assert internal metrics the agent chose.\n- **An input is copied but never consumed.** A directory or file named in the task appears in the deliverable solely as a copy step. Fix: Pass A — list every named input and bind each to a downstream use; add a verification assertion per input.\n- **Cross-file lookup fails on case-sensitive filesystem.** Identifier casing in one artifact (e.g., ticker column `SPY`) does not match filename casing in a sibling data directory (e.g., `spy.csv`). The same mechanism breaks any lookup across whitespace, punctuation, or encoding drift between artifacts. Fix: normalize one side so every identifier resolves; assert the lookup for every id before writing.\n- **Table re-sorted, decorated, or placeholder-invented.** Calling `.sort()`, wrapping cells in `\u003Cspan>`\u002F`\u003Cdiv>`, or formatting missing values as `\"—\"` \u002F `\"N\u002FA\"` \u002F `\"-\"` when the task is silent on the subset — all break strict evaluators. Fix: identity transformation — file order, plain text, empty cell when the task is silent and its example shows a non-empty format for the column; mirror the task's placeholder string only when it names one.\n- **Tooltip suppressed on a subset the task said \"lacks\" some fields.** Removing mouse handlers or the carrier entirely for ETFs\u002Findexes\u002Fetc. deletes a DOM feature the task expected on every mark. Fix: attach the same single carrier to every mark; omit missing fields from the text, do not label them `\"N\u002FA\"` unless the task uses that string.\n- **Invented wrapper DOM nesting.** Wrapping each mark in a `\u003Cg class=\"…-group\">` and moving `data-*` identifiers onto the wrapper rather than the mark itself fails evaluators that query `circle[data-ticker]`. Fix: keep mark + label as flat siblings; put `data-*` on the mark.\n",{"name":5,"description":7},"main","cdcb6d230f3b15c3c33f8a2bdd16439f43602cbe","evaluation\u002Fresults\u002FSkillGrad\u002Fopus47_clean_20261016_partial14\u002Ftasks\u002Fdata-to-d3\u002Frepaired_skill\u002Fd3-visualization",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 d3-visualization",{"tool":37,"command":38},"bunx","bunx skills add linyouxue\u002Fskill-repair-benchmark --skill d3-visualization",{"tool":40,"command":41},"pnpm","pnpm dlx skills add linyouxue\u002Fskill-repair-benchmark --skill d3-visualization","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Flinyouxue\u002Fskill-repair-benchmark\u002Fblob\u002Fmain\u002Fevaluation\u002Fresults\u002FSkillGrad\u002Fopus47_clean_20261016_partial14\u002Ftasks\u002Fdata-to-d3\u002Frepaired_skill\u002Fd3-visualization\u002FSKILL.md",true,[45,51,55,59],{"path":46,"resourceType":47,"resourceTypeDesc":48,"sizeBytes":49,"sourceHtmlUrl":50},"scripts\u002Fbubble_chart_example.js","SCRIPT","脚本",3679,"https:\u002F\u002Fgithub.com\u002Flinyouxue\u002Fskill-repair-benchmark\u002Fblob\u002Fmain\u002Fevaluation\u002Fresults\u002FSkillGrad\u002Fopus47_clean_20261016_partial14\u002Ftasks\u002Fdata-to-d3\u002Frepaired_skill\u002Fd3-visualization\u002Fscripts\u002Fbubble_chart_example.js",{"path":52,"resourceType":47,"resourceTypeDesc":48,"sizeBytes":53,"sourceHtmlUrl":54},"scripts\u002Fcheck_tooltip.js",1752,"https:\u002F\u002Fgithub.com\u002Flinyouxue\u002Fskill-repair-benchmark\u002Fblob\u002Fmain\u002Fevaluation\u002Fresults\u002FSkillGrad\u002Fopus47_clean_20261016_partial14\u002Ftasks\u002Fdata-to-d3\u002Frepaired_skill\u002Fd3-visualization\u002Fscripts\u002Fcheck_tooltip.js",{"path":56,"resourceType":47,"resourceTypeDesc":48,"sizeBytes":57,"sourceHtmlUrl":58},"scripts\u002Finteractive_table_example.js",4439,"https:\u002F\u002Fgithub.com\u002Flinyouxue\u002Fskill-repair-benchmark\u002Fblob\u002Fmain\u002Fevaluation\u002Fresults\u002FSkillGrad\u002Fopus47_clean_20261016_partial14\u002Ftasks\u002Fdata-to-d3\u002Frepaired_skill\u002Fd3-visualization\u002Fscripts\u002Finteractive_table_example.js",{"path":60,"resourceType":47,"resourceTypeDesc":48,"sizeBytes":61,"sourceHtmlUrl":62},"scripts\u002Ftooltip_handler.js",5741,"https:\u002F\u002Fgithub.com\u002Flinyouxue\u002Fskill-repair-benchmark\u002Fblob\u002Fmain\u002Fevaluation\u002Fresults\u002FSkillGrad\u002Fopus47_clean_20261016_partial14\u002Ftasks\u002Fdata-to-d3\u002Frepaired_skill\u002Fd3-visualization\u002Fscripts\u002Ftooltip_handler.js",[64,71,78,85,90,95],{"skillNo":65,"name":66,"nameZh":6,"description":67,"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},"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":72,"name":73,"nameZh":6,"description":74,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":75,"categories":76,"keywords":77,"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":79,"name":80,"nameZh":6,"description":81,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":82,"categories":83,"keywords":84,"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":86,"name":66,"nameZh":6,"description":67,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":87,"categories":88,"keywords":89,"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":91,"name":73,"nameZh":6,"description":74,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":92,"categories":93,"keywords":94,"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":96,"name":80,"nameZh":6,"description":81,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":97,"categories":98,"keywords":99,"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",[],[]]