[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M752ZK7JGW7P03NT4W2WKBKB":3,"skill-same:01M752ZK7JGW7P03NT4W2WKBKB":52},{"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":28,"commitSha":29,"skillDirectory":30,"forksCount":31,"repositoryHtmlUrl":32,"repositoryHomepageUrl":33,"ownerHtmlUrl":34,"ownerBlogUrl":35,"ownerDescription":36,"updatedAt":37,"install":38,"resources":51},"01M752ZK7JGW7P03NT4W2WKBKB","cv-import",null,"Extract a PDF\u002FDOCX CV, enhance it with LLM, and transform it into the qwik-resume-editor JSON schema (Resume v2.0.0). Uses markitdown if available, falls back to LLM extraction.","5kahoisaac\u002Fresume","5kahoisaac","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F49708270?v=4","01M752ZDYNS5X656FH0ZMZRW9Y","Isaac","01M752ZDZ84X56YFD4V761Z9GG",1,"MIT","https:\u002F\u002Fgithub.com\u002F5kahoisaac\u002Fresume\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fcv-import\u002FSKILL.md",[],[],"2026-11-12 13:55:43",false,"---\nname: cv-import\ndescription: \"Extract a PDF\u002FDOCX CV, enhance it with LLM, and transform it into the qwik-resume-editor JSON schema (Resume v2.0.0). Uses markitdown if available, falls back to LLM extraction.\"\nallowed-tools: [ Read, Write, Bash ]\n---\n\n# CV Import — PDF to Resume JSON\n\nImport any CV\u002Fresume PDF into the qwik-resume-editor JSON format in three stages:\n**Extract → Enhance → Transform**.\n\n## Usage\n\n```\n\u002Fcv-import \u002Fpath\u002Fto\u002Fcv.pdf                  # writes src\u002Fdata\u002Fdefault-resume.json (default)\n\u002Fcv-import \u002Fpath\u002Fto\u002Fcv.pdf --output resume.json   # write to a custom file instead\n```\n\n## Stage 0 — Validate Input\n\n1. Confirm the file path was provided. If not, ask the user: `\"Please provide the path to your CV PDF.\"`\n2. Check the file exists:\n   ```bash\n   test -f \"\u003Cfilepath>\" && echo \"exists\" || echo \"not found\"\n   ```\n3. Accept `.pdf`, `.docx`, `.doc`, `.txt`. Warn (but proceed) for other extensions.\n\n## Stage 1 — Extract Text\n\n### 1a. Check for markitdown\n\n```bash\nmarkitdown --version 2>\u002Fdev\u002Fnull && echo \"MARKITDOWN_OK\" || echo \"MARKITDOWN_ABSENT\"\n```\n\n### 1b-A. If markitdown is available\n\n```bash\nmarkitdown \"\u003Cfilepath>\"\n```\n\nCapture the full stdout as the extracted markdown. markitdown handles multi-column PDFs and strips headers\u002Ffooters well.\n\n### 1b-B. If markitdown is absent — LLM extraction\n\nUse the `Read` tool to open the PDF directly. Claude can read PDF files natively. If that fails (non-text PDF), instruct\nthe user:\n\n```\nmarkitdown is not installed. Install it for best results:\n  pip install markitdown\n  # or: pipx install markitdown\n\nFalling back to native PDF reading via Claude…\n```\n\nAfter reading, extract **all visible text**, preserving approximate section order. Do not summarise at this stage — get\neverything.\n\n## Stage 2 — Enhance\n\nAfter extraction, use LLM reasoning to:\n\n1. **Identify sections** — find headings that correspond to: Summary\u002FObjective, Experience\u002FWork History, Education,\n   Skills, Languages, Certifications, Awards\u002FHonors, References.\n2. **Normalise dates** — convert any format (`Jan 2020`, `01\u002F2020`, `2020`) to `\"YYYY-MM\"`. Use `\"\"` (empty string)\n   for \"Present\" \u002F \"Current\" \u002F \"Now\".\n3. **Clean description text** — convert bullet points to HTML `\u003Cul>\u003Cli>…\u003C\u002Fli>\u003C\u002Ful>`. Wrap paragraphs in `\u003Cp>…\u003C\u002Fp>`.\n   Preserve bold\u002Fitalic if detectable.\n4. **Infer contact types** — classify each contact detail:\n    - email address → `type: \"email\"`\n    - phone number → `type: \"tel\"`\n    - URL (http\u002Fhttps, LinkedIn, GitHub, portfolio) → `type: \"url\"`\n    - plain text (city, country) → `type: \"string\"`\n5. **Infer language levels** — map free-text proficiency to the enum:\n   `Basic | Elementary | Intermediate | Conversational | Proficient | Advanced | Native`\n6. **Score expertise** — if expertise\u002Fcompetency levels are present, map to 0–100.\n7. **Flag uncertain fields** — add a `_review` key next to uncertain values with a note, e.g.\n   `\"_review_start\": \"date was '2020' — assumed January\"`.\n\n## Stage 3 — Transform to Resume JSON\n\nProduce a valid `Resume` object (schema version `\"2.0.0\"`).\n\n### ID generation\n\nUse this pattern for every `id` field: `\u003Cprefix>_\u003C7 alphanumeric chars>`.\nGenerate sequential IDs: `c_0000001`, `sec_0000001`, `exp_0000001`, `edu_0000001`, `grp_0000001`, `lng_0000001`,\n`cer_0000001`, `awd_0000001`, `ref_0000001`.\n\n### JSON schema\n\n```jsonc\n{\n  \"version\": \"2.0.0\",\n  \"header\": {\n    \"name\": \"Full Name\",\n    \"title\": \"Professional Title \u002F Headline\",\n    \"contacts\": [\n      \u002F\u002F Order: email, phone, LinkedIn, website, location last\n      { \"id\": \"c_0000001\", \"type\": \"email\",  \"label\": \"you@example.com\" },\n      { \"id\": \"c_0000002\", \"type\": \"tel\",    \"label\": \"+1 234 567 8900\" },\n      { \"id\": \"c_0000003\", \"type\": \"url\",    \"label\": \"linkedin.com\u002Fin\u002Fyou\", \"href\": \"https:\u002F\u002Flinkedin.com\u002Fin\u002Fyou\" },\n      { \"id\": \"c_0000004\", \"type\": \"string\", \"label\": \"City, Country\" }\n    ]\n  },\n  \"theme\": { \"paletteId\": \"orange-navy\" },\n  \"sections\": [\n    {\n      \"id\": \"sec_0000001\",\n      \"type\": \"summary\",\n      \"title\": \"Summary\",\n      \"visible\": true,\n      \"data\": { \"text\": \"\u003Cp>Professional summary as HTML…\u003C\u002Fp>\" }\n    },\n    {\n      \"id\": \"sec_0000002\",\n      \"type\": \"experience\",\n      \"title\": \"Experience\",\n      \"visible\": true,\n      \"data\": {\n        \"items\": [\n          {\n            \"id\": \"exp_0000001\",\n            \"title\": \"Job Title\",\n            \"company\": \"Company Name\",\n            \"location\": \"City, Country\",\n            \"start\": \"2021-03\",\n            \"end\": \"\",\n            \"description\": \"\u003Cul>\u003Cli>Achievement or responsibility\u003C\u002Fli>\u003C\u002Ful>\"\n          }\n        ]\n      }\n    },\n    {\n      \"id\": \"sec_0000003\",\n      \"type\": \"education\",\n      \"title\": \"Education\",\n      \"visible\": true,\n      \"data\": {\n        \"items\": [\n          {\n            \"id\": \"edu_0000001\",\n            \"degree\": \"Bachelor of Science in Computer Science\",\n            \"school\": \"University Name\",\n            \"start\": \"2015-09\",\n            \"end\": \"2019-06\"\n          }\n        ]\n      }\n    },\n    {\n      \"id\": \"sec_0000004\",\n      \"type\": \"skills\",\n      \"title\": \"Skills & Knowledge\",\n      \"visible\": true,\n      \"data\": {\n        \"groups\": [\n          {\n            \"id\": \"grp_0000001\",\n            \"label\": \"Programming Languages\",\n            \"skills\": [\"Python\", \"TypeScript\", \"Go\"]\n          }\n        ]\n      }\n    },\n    {\n      \"id\": \"sec_0000005\",\n      \"type\": \"languages\",\n      \"title\": \"Languages\",\n      \"visible\": true,\n      \"data\": {\n        \"items\": [\n          { \"id\": \"lng_0000001\", \"name\": \"English\", \"level\": \"Native\" },\n          { \"id\": \"lng_0000002\", \"name\": \"Mandarin\", \"level\": \"Conversational\" }\n        ]\n      }\n    },\n    {\n      \"id\": \"sec_0000006\",\n      \"type\": \"certifications\",\n      \"title\": \"Licenses & Certifications\",\n      \"visible\": true,\n      \"data\": {\n        \"items\": [\n          { \"id\": \"cer_0000001\", \"name\": \"AWS Solutions Architect\", \"issuer\": \"Amazon Web Services\" }\n        ]\n      }\n    }\n    \u002F\u002F Only include section types present in the CV.\n    \u002F\u002F Available types: summary | experience | education | skills | languages |\n    \u002F\u002F                  certifications | awards | expertise | references\n  ]\n}\n```\n\n### Section ordering heuristic\n\nDefault order (omit sections not found in the CV):\n\n1. `summary`\n2. `experience`\n3. `education`\n4. `skills`\n5. `expertise`\n6. `languages`\n7. `certifications`\n8. `awards`\n9. `references`\n\n### Skills grouping heuristic\n\nIf the CV lists skills as a flat list, group by inferred category:\n\n- Programming Languages\n- Frameworks & Libraries\n- Tools & Platforms\n- Databases\n- Soft Skills \u002F Other\n\nIf no clear grouping exists, use a single group `\"label\": \"Skills\"`.\n\n### Expertise section\n\nOnly emit an `expertise` section if the CV explicitly shows percentage bars, star ratings, or numbered proficiency\nlevels. Map to 0–100:\n\n- 5\u002F5 stars → 100, 4\u002F5 → 80, 3\u002F5 → 60, 2\u002F5 → 40, 1\u002F5 → 20\n- Percentages: use directly.\n\n```jsonc\n{\n  \"id\": \"sec_000000X\",\n  \"type\": \"expertise\",\n  \"title\": \"Industry Expertise\",\n  \"visible\": true,\n  \"data\": {\n    \"items\": [\n      { \"id\": \"xp_0000001\", \"label\": \"Machine Learning\", \"level\": 80 }\n    ]\n  }\n}\n```\n\n## Stage 4 — Output\n\n1. **Write `src\u002Fdata\u002Fdefault-resume.json`** in the project root (default behaviour). This is the file the app bundles as\n   the seed shown on first visit and on Reset — no editor login required to see the result.\n    - If `--output \u003Cfile>` was specified, write to that path instead.\n2. **Print the full JSON** to the conversation so the user can review it.\n3. **Summarise what was imported** in a short table:\n\n| Field            | Value                                                    |\n|------------------|----------------------------------------------------------|\n| Name             | …                                                        |\n| Sections found   | experience (N jobs), education (N), skills (N groups), … |\n| Dates normalised | N                                                        |\n| Fields to review | N (marked with `_review_*` keys)                         |\n\n4. **Next steps** — tell the user exactly these two steps to preview the resume without logging in:\n\n   Open `http:\u002F\u002Flocalhost:5173` in the browser, open DevTools console (F12 \u002F Cmd+Option+I), paste this one line, and\n   press Enter:\n   ```js\n   localStorage.removeItem(\"qwik-resume-editor:v2\"); location.reload();\n   ```\n   This clears any cached resume from localStorage so the app falls back to the newly written `default-resume.json`. The\n   page reloads and shows the imported resume on the public home page — no password needed.\n\n## Error handling\n\n| Problem                     | Action                                                                                                                        |\n|-----------------------------|-------------------------------------------------------------------------------------------------------------------------------|\n| File not found              | Report path, ask user to confirm                                                                                              |\n| PDF is scanned \u002F image-only | Warn \"no text layer detected\"; suggest installing markitdown with OCR support (`pip install markitdown[pdf]`) or manual entry |\n| Date cannot be parsed       | Use `\"\"` and add `\"_review_date\": \"original: \u003Craw text>\"`                                                                     |\n| Section heading ambiguous   | Pick the closest match; add `\"_review_section\": \"guessed from: \u003Cheading text>\"`                                               |\n| markitdown exits with error | Fall back to LLM extraction; report the error message                                                                         |\n| Contact type unclear        | Default to `\"string\"` type; note it in the summary                                                                            |\n\n## Quality checks before output\n\n- [ ] All `id` fields are unique within the document\n- [ ] All dates are `\"YYYY-MM\"` or `\"\"` — no other formats\n- [ ] `type` fields match the allowed `SectionType` enum exactly\n- [ ] `level` fields in `languages` use the exact enum values:\n  `Basic | Elementary | Intermediate | Conversational | Proficient | Advanced | Native`\n- [ ] `level` in `expertise` items is a number 0–100\n- [ ] `description` fields contain HTML, not plain text with hyphens\n- [ ] No section type appears more than once in `sections`\n- [ ] `version` is exactly `\"2.0.0\"`\n- [ ] `theme.paletteId` is one of: `orange-navy | navy-black | green-charcoal | purple-slate | magenta-graphite | mono`\n- [ ] JSON is valid (no trailing commas, no `\u002F\u002F` comments in the actual output)\n","# CV Import — PDF to Resume JSON\n\nImport any CV\u002Fresume PDF into the qwik-resume-editor JSON format in three stages:\n**Extract → Enhance → Transform**.\n\n## Usage\n\n```\n\u002Fcv-import \u002Fpath\u002Fto\u002Fcv.pdf                  # writes src\u002Fdata\u002Fdefault-resume.json (default)\n\u002Fcv-import \u002Fpath\u002Fto\u002Fcv.pdf --output resume.json   # write to a custom file instead\n```\n\n## Stage 0 — Validate Input\n\n1. Confirm the file path was provided. If not, ask the user: `\"Please provide the path to your CV PDF.\"`\n2. Check the file exists:\n   ```bash\n   test -f \"\u003Cfilepath>\" && echo \"exists\" || echo \"not found\"\n   ```\n3. Accept `.pdf`, `.docx`, `.doc`, `.txt`. Warn (but proceed) for other extensions.\n\n## Stage 1 — Extract Text\n\n### 1a. Check for markitdown\n\n```bash\nmarkitdown --version 2>\u002Fdev\u002Fnull && echo \"MARKITDOWN_OK\" || echo \"MARKITDOWN_ABSENT\"\n```\n\n### 1b-A. If markitdown is available\n\n```bash\nmarkitdown \"\u003Cfilepath>\"\n```\n\nCapture the full stdout as the extracted markdown. markitdown handles multi-column PDFs and strips headers\u002Ffooters well.\n\n### 1b-B. If markitdown is absent — LLM extraction\n\nUse the `Read` tool to open the PDF directly. Claude can read PDF files natively. If that fails (non-text PDF), instruct\nthe user:\n\n```\nmarkitdown is not installed. Install it for best results:\n  pip install markitdown\n  # or: pipx install markitdown\n\nFalling back to native PDF reading via Claude…\n```\n\nAfter reading, extract **all visible text**, preserving approximate section order. Do not summarise at this stage — get\neverything.\n\n## Stage 2 — Enhance\n\nAfter extraction, use LLM reasoning to:\n\n1. **Identify sections** — find headings that correspond to: Summary\u002FObjective, Experience\u002FWork History, Education,\n   Skills, Languages, Certifications, Awards\u002FHonors, References.\n2. **Normalise dates** — convert any format (`Jan 2020`, `01\u002F2020`, `2020`) to `\"YYYY-MM\"`. Use `\"\"` (empty string)\n   for \"Present\" \u002F \"Current\" \u002F \"Now\".\n3. **Clean description text** — convert bullet points to HTML `\u003Cul>\u003Cli>…\u003C\u002Fli>\u003C\u002Ful>`. Wrap paragraphs in `\u003Cp>…\u003C\u002Fp>`.\n   Preserve bold\u002Fitalic if detectable.\n4. **Infer contact types** — classify each contact detail:\n    - email address → `type: \"email\"`\n    - phone number → `type: \"tel\"`\n    - URL (http\u002Fhttps, LinkedIn, GitHub, portfolio) → `type: \"url\"`\n    - plain text (city, country) → `type: \"string\"`\n5. **Infer language levels** — map free-text proficiency to the enum:\n   `Basic | Elementary | Intermediate | Conversational | Proficient | Advanced | Native`\n6. **Score expertise** — if expertise\u002Fcompetency levels are present, map to 0–100.\n7. **Flag uncertain fields** — add a `_review` key next to uncertain values with a note, e.g.\n   `\"_review_start\": \"date was '2020' — assumed January\"`.\n\n## Stage 3 — Transform to Resume JSON\n\nProduce a valid `Resume` object (schema version `\"2.0.0\"`).\n\n### ID generation\n\nUse this pattern for every `id` field: `\u003Cprefix>_\u003C7 alphanumeric chars>`.\nGenerate sequential IDs: `c_0000001`, `sec_0000001`, `exp_0000001`, `edu_0000001`, `grp_0000001`, `lng_0000001`,\n`cer_0000001`, `awd_0000001`, `ref_0000001`.\n\n### JSON schema\n\n```jsonc\n{\n  \"version\": \"2.0.0\",\n  \"header\": {\n    \"name\": \"Full Name\",\n    \"title\": \"Professional Title \u002F Headline\",\n    \"contacts\": [\n      \u002F\u002F Order: email, phone, LinkedIn, website, location last\n      { \"id\": \"c_0000001\", \"type\": \"email\",  \"label\": \"you@example.com\" },\n      { \"id\": \"c_0000002\", \"type\": \"tel\",    \"label\": \"+1 234 567 8900\" },\n      { \"id\": \"c_0000003\", \"type\": \"url\",    \"label\": \"linkedin.com\u002Fin\u002Fyou\", \"href\": \"https:\u002F\u002Flinkedin.com\u002Fin\u002Fyou\" },\n      { \"id\": \"c_0000004\", \"type\": \"string\", \"label\": \"City, Country\" }\n    ]\n  },\n  \"theme\": { \"paletteId\": \"orange-navy\" },\n  \"sections\": [\n    {\n      \"id\": \"sec_0000001\",\n      \"type\": \"summary\",\n      \"title\": \"Summary\",\n      \"visible\": true,\n      \"data\": { \"text\": \"\u003Cp>Professional summary as HTML…\u003C\u002Fp>\" }\n    },\n    {\n      \"id\": \"sec_0000002\",\n      \"type\": \"experience\",\n      \"title\": \"Experience\",\n      \"visible\": true,\n      \"data\": {\n        \"items\": [\n          {\n            \"id\": \"exp_0000001\",\n            \"title\": \"Job Title\",\n            \"company\": \"Company Name\",\n            \"location\": \"City, Country\",\n            \"start\": \"2021-03\",\n            \"end\": \"\",\n            \"description\": \"\u003Cul>\u003Cli>Achievement or responsibility\u003C\u002Fli>\u003C\u002Ful>\"\n          }\n        ]\n      }\n    },\n    {\n      \"id\": \"sec_0000003\",\n      \"type\": \"education\",\n      \"title\": \"Education\",\n      \"visible\": true,\n      \"data\": {\n        \"items\": [\n          {\n            \"id\": \"edu_0000001\",\n            \"degree\": \"Bachelor of Science in Computer Science\",\n            \"school\": \"University Name\",\n            \"start\": \"2015-09\",\n            \"end\": \"2019-06\"\n          }\n        ]\n      }\n    },\n    {\n      \"id\": \"sec_0000004\",\n      \"type\": \"skills\",\n      \"title\": \"Skills & Knowledge\",\n      \"visible\": true,\n      \"data\": {\n        \"groups\": [\n          {\n            \"id\": \"grp_0000001\",\n            \"label\": \"Programming Languages\",\n            \"skills\": [\"Python\", \"TypeScript\", \"Go\"]\n          }\n        ]\n      }\n    },\n    {\n      \"id\": \"sec_0000005\",\n      \"type\": \"languages\",\n      \"title\": \"Languages\",\n      \"visible\": true,\n      \"data\": {\n        \"items\": [\n          { \"id\": \"lng_0000001\", \"name\": \"English\", \"level\": \"Native\" },\n          { \"id\": \"lng_0000002\", \"name\": \"Mandarin\", \"level\": \"Conversational\" }\n        ]\n      }\n    },\n    {\n      \"id\": \"sec_0000006\",\n      \"type\": \"certifications\",\n      \"title\": \"Licenses & Certifications\",\n      \"visible\": true,\n      \"data\": {\n        \"items\": [\n          { \"id\": \"cer_0000001\", \"name\": \"AWS Solutions Architect\", \"issuer\": \"Amazon Web Services\" }\n        ]\n      }\n    }\n    \u002F\u002F Only include section types present in the CV.\n    \u002F\u002F Available types: summary | experience | education | skills | languages |\n    \u002F\u002F                  certifications | awards | expertise | references\n  ]\n}\n```\n\n### Section ordering heuristic\n\nDefault order (omit sections not found in the CV):\n\n1. `summary`\n2. `experience`\n3. `education`\n4. `skills`\n5. `expertise`\n6. `languages`\n7. `certifications`\n8. `awards`\n9. `references`\n\n### Skills grouping heuristic\n\nIf the CV lists skills as a flat list, group by inferred category:\n\n- Programming Languages\n- Frameworks & Libraries\n- Tools & Platforms\n- Databases\n- Soft Skills \u002F Other\n\nIf no clear grouping exists, use a single group `\"label\": \"Skills\"`.\n\n### Expertise section\n\nOnly emit an `expertise` section if the CV explicitly shows percentage bars, star ratings, or numbered proficiency\nlevels. Map to 0–100:\n\n- 5\u002F5 stars → 100, 4\u002F5 → 80, 3\u002F5 → 60, 2\u002F5 → 40, 1\u002F5 → 20\n- Percentages: use directly.\n\n```jsonc\n{\n  \"id\": \"sec_000000X\",\n  \"type\": \"expertise\",\n  \"title\": \"Industry Expertise\",\n  \"visible\": true,\n  \"data\": {\n    \"items\": [\n      { \"id\": \"xp_0000001\", \"label\": \"Machine Learning\", \"level\": 80 }\n    ]\n  }\n}\n```\n\n## Stage 4 — Output\n\n1. **Write `src\u002Fdata\u002Fdefault-resume.json`** in the project root (default behaviour). This is the file the app bundles as\n   the seed shown on first visit and on Reset — no editor login required to see the result.\n    - If `--output \u003Cfile>` was specified, write to that path instead.\n2. **Print the full JSON** to the conversation so the user can review it.\n3. **Summarise what was imported** in a short table:\n\n| Field            | Value                                                    |\n|------------------|----------------------------------------------------------|\n| Name             | …                                                        |\n| Sections found   | experience (N jobs), education (N), skills (N groups), … |\n| Dates normalised | N                                                        |\n| Fields to review | N (marked with `_review_*` keys)                         |\n\n4. **Next steps** — tell the user exactly these two steps to preview the resume without logging in:\n\n   Open `http:\u002F\u002Flocalhost:5173` in the browser, open DevTools console (F12 \u002F Cmd+Option+I), paste this one line, and\n   press Enter:\n   ```js\n   localStorage.removeItem(\"qwik-resume-editor:v2\"); location.reload();\n   ```\n   This clears any cached resume from localStorage so the app falls back to the newly written `default-resume.json`. The\n   page reloads and shows the imported resume on the public home page — no password needed.\n\n## Error handling\n\n| Problem                     | Action                                                                                                                        |\n|-----------------------------|-------------------------------------------------------------------------------------------------------------------------------|\n| File not found              | Report path, ask user to confirm                                                                                              |\n| PDF is scanned \u002F image-only | Warn \"no text layer detected\"; suggest installing markitdown with OCR support (`pip install markitdown[pdf]`) or manual entry |\n| Date cannot be parsed       | Use `\"\"` and add `\"_review_date\": \"original: \u003Craw text>\"`                                                                     |\n| Section heading ambiguous   | Pick the closest match; add `\"_review_section\": \"guessed from: \u003Cheading text>\"`                                               |\n| markitdown exits with error | Fall back to LLM extraction; report the error message                                                                         |\n| Contact type unclear        | Default to `\"string\"` type; note it in the summary                                                                            |\n\n## Quality checks before output\n\n- [ ] All `id` fields are unique within the document\n- [ ] All dates are `\"YYYY-MM\"` or `\"\"` — no other formats\n- [ ] `type` fields match the allowed `SectionType` enum exactly\n- [ ] `level` fields in `languages` use the exact enum values:\n  `Basic | Elementary | Intermediate | Conversational | Proficient | Advanced | Native`\n- [ ] `level` in `expertise` items is a number 0–100\n- [ ] `description` fields contain HTML, not plain text with hyphens\n- [ ] No section type appears more than once in `sections`\n- [ ] `version` is exactly `\"2.0.0\"`\n- [ ] `theme.paletteId` is one of: `orange-navy | navy-black | green-charcoal | purple-slate | magenta-graphite | mono`\n- [ ] JSON is valid (no trailing commas, no `\u002F\u002F` comments in the actual output)\n",{"name":5,"description":7,"allowed-tools":24},[25,26,27],"Read","Write","Bash","main","c31a9389f9f97f4cbb2787f590b375436e6a799a",".claude\u002Fskills\u002Fcv-import",0,"https:\u002F\u002Fgithub.com\u002F5kahoisaac\u002Fresume","https:\u002F\u002Fresume.isaac.ng","https:\u002F\u002Fgithub.com\u002F5kahoisaac","https:\u002F\u002Fisaac.ng","A random programmer","2026-08-17 11:09:20",{"source":8,"skillName":5,"commands":39,"prompt":49,"zipAvailable":50},[40,43,46],{"tool":41,"command":42},"npx","npx skills add 5kahoisaac\u002Fresume --skill cv-import",{"tool":44,"command":45},"bunx","bunx skills add 5kahoisaac\u002Fresume --skill cv-import",{"tool":47,"command":48},"pnpm","pnpm dlx skills add 5kahoisaac\u002Fresume --skill cv-import","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002F5kahoisaac\u002Fresume\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fcv-import\u002FSKILL.md",true,[],[53,58,63,68,73,78],{"skillNo":54,"name":5,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":55,"categories":56,"keywords":57,"publishedAt":19},"01M752ZSE35RCYT3R48RH4JF9H","https:\u002F\u002Fgithub.com\u002F5kahoisaac\u002Fresume\u002Fblob\u002Fmain\u002F.pi\u002Fskills\u002Fcv-import\u002FSKILL.md",[],[],{"skillNo":59,"name":5,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":60,"categories":61,"keywords":62,"publishedAt":19},"01M752ZRNFYGGPYBCGFAYGQVB5","https:\u002F\u002Fgithub.com\u002F5kahoisaac\u002Fresume\u002Fblob\u002Fmain\u002F.opencode\u002Fskills\u002Fcv-import\u002FSKILL.md",[],[],{"skillNo":64,"name":5,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":65,"categories":66,"keywords":67,"publishedAt":19},"01M752ZQQGS2GSSPPB17YKK7W1","https:\u002F\u002Fgithub.com\u002F5kahoisaac\u002Fresume\u002Fblob\u002Fmain\u002F.gemini\u002Fskills\u002Fcv-import\u002FSKILL.md",[],[],{"skillNo":69,"name":5,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":70,"categories":71,"keywords":72,"publishedAt":19},"01M752ZPHJT4YDCQNN8JYFMAS2","https:\u002F\u002Fgithub.com\u002F5kahoisaac\u002Fresume\u002Fblob\u002Fmain\u002F.cursor\u002Fskills\u002Fcv-import\u002FSKILL.md",[],[],{"skillNo":74,"name":5,"nameZh":6,"description":6,"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},"01M752ZNPHSVZG4KQ27Y8VDVJK","https:\u002F\u002Fgithub.com\u002F5kahoisaac\u002Fresume\u002Fblob\u002Fmain\u002F.codex\u002Fskills\u002Fcv-import\u002FSKILL.md",[],[],{"skillNo":79,"name":5,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":80,"categories":81,"keywords":82,"publishedAt":19},"01M752ZH968DFRHXDNNS5F5E3F","https:\u002F\u002Fgithub.com\u002F5kahoisaac\u002Fresume\u002Fblob\u002Fmain\u002F.antigravity\u002Fskills\u002Fcv-import\u002FSKILL.md",[],[]]