[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7AWEWWKPR95HJ447YK1NTM2":3,"skill-same:01M7AWEWWKPR95HJ447YK1NTM2":47},{"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":30,"ownerDescription":31,"updatedAt":32,"install":33,"resources":46},"01M7AWEWWKPR95HJ447YK1NTM2","jobsearch",null,"Search job listings in the US (USAJOBS federal, Adzuna private market, and Greenhouse\u002FLever\u002FAshby company boards) and Denmark (Jobindex, Jobnet, Akademikernes Jobbank, Jobdanmark), then run the AI job-application workflow from the ai-job-search repo. Use when the user says \"\u002Fjobsearch\", \"find jobs\", \"search jobs\", \"jobs in USA \u002F [US city]\", \"federal jobs\", \"USAJOBS\", \"Adzuna\", \"jobs at [company]\", \"Danish jobs\", \"jobindex\u002Fjobnet\u002Fjobbank\u002Fjobdanmark\", wants to look up listings, job categories, occupation codes, or salary benchmarks, or wants to evaluate fit \u002F tailor a CV \u002F write a cover letter for a posting. Wraps nine Bun CLIs (5 US sources + 4 Danish portals) plus the repo's \u002Fsetup, \u002Fscrape, \u002Fapply commands.","barkleesanders\u002Fclaude-code-starter","barkleesanders","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F30680901?v=4","01M7AW019CGBYWDA8MFJEN1JXC","Barklee Sanders","01M7AW01A5H994ASJ2QZPW67AV",4,"MIT","https:\u002F\u002Fgithub.com\u002Fbarkleesanders\u002Fclaude-code-starter\u002Fblob\u002Fmain\u002Fskills\u002Fjobsearch\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: jobsearch\ndescription: Search job listings in the US (USAJOBS federal, Adzuna private market, and Greenhouse\u002FLever\u002FAshby company boards) and Denmark (Jobindex, Jobnet, Akademikernes Jobbank, Jobdanmark), then run the AI job-application workflow from the ai-job-search repo. Use when the user says \"\u002Fjobsearch\", \"find jobs\", \"search jobs\", \"jobs in USA \u002F [US city]\", \"federal jobs\", \"USAJOBS\", \"Adzuna\", \"jobs at [company]\", \"Danish jobs\", \"jobindex\u002Fjobnet\u002Fjobbank\u002Fjobdanmark\", wants to look up listings, job categories, occupation codes, or salary benchmarks, or wants to evaluate fit \u002F tailor a CV \u002F write a cover letter for a posting. Wraps nine Bun CLIs (5 US sources + 4 Danish portals) plus the repo's \u002Fsetup, \u002Fscrape, \u002Fapply commands.\n---\n\n# \u002Fjobsearch — US + Danish Job Search + Application Assistant\n\nThin launcher over the job-portal CLIs in the **ai-job-search** repo. **Default to the US sources** unless the user asks for Denmark. Each CLI emits clean JSON to stdout; errors are JSON on stderr with exit 1. Same run rule everywhere: **`cd` into the CLI's `cli\u002F` dir first** (bun resolves `node_modules` by file location), and never bump `@bunli\u002Fcore` past `0.7.0`.\n\n---\n\n## 🇺🇸 US jobs — `us-jobs` CLI\n\n```bash\ncd ~\u002Fai-job-search\u002F.agents\u002Fskills\u002Fus-jobs\u002Fcli && bun run src\u002Fcli.ts \u003Ccommand> [flags]\n```\n\nFive sources, one normalized output shape: `{meta, results:[{source,id,title,company,location,remote,salary,url,posted}]}`. All accept `--limit` and `--format json|table|plain`.\n\n### Keyless — company job boards (work immediately, ✅ verified live)\nFor \"jobs at \u003Ccompany>\", the **default is `auto`** — it fans out across all three ATSes in parallel so you don't have to know which one the company uses:\n```bash\nbun run src\u002Fcli.ts auto \u003Ccompany> -q engineer --remote --limit 20   # tries greenhouse+lever+ashby, merges, dedupes\n```\n`meta.sources_hit` tells you which ATS actually had the company. `-q` filters by title text. Use the single-source commands only when you already know the ATS:\n```bash\nbun run src\u002Fcli.ts greenhouse \u003Cboard>   --limit 20   # e.g. stripe, airbnb, coinbase\nbun run src\u002Fcli.ts lever      \u003Caccount>  --limit 20   # e.g. palantir\nbun run src\u002Fcli.ts ashby      \u003Cboard>    --remote     # e.g. Notion, Ramp, linear, openai\n```\n`\u003Ccompany>`\u002F`\u003Cboard>`\u002F`\u003Caccount>` is the company's slug in its careers URL (`jobs.lever.co\u002F\u003Caccount>`, `jobs.ashbyhq.com\u002F\u003Cboard>`, `boards.greenhouse.io\u002F\u003Cboard>`). `NOT_FOUND` from `auto` means the company isn't on any of the three keyless ATSes (try Adzuna, or check their careers page for a different ATS).\n\n### Federal — USAJOBS (✅ key configured, verified live)\n```bash\nbun run src\u002Fcli.ts usajobs -q \"data scientist\" --location \"Austin, TX\" --remote --limit 20\n```\n\n### Broad private market — Adzuna (✅ key configured, verified live)\n```bash\nbun run src\u002Fcli.ts adzuna -q \"software engineer\" --where \"Seattle, WA\" --salary-min 120000 --remote --limit 20\n```\n\n**Keys are already set up** (2026-06-04) in `us-jobs\u002Fcli\u002F.env` — bun auto-loads it, so commands just work from the cli dir with no `export` needed. The same keys are also in `~\u002F.zshrc` for interactive terminals. `.env` is gitignored (keys never commit). If you ever see a `MISSING_KEY` error, the `.env` was moved\u002Flost — re-add `ADZUNA_APP_ID`\u002F`ADZUNA_APP_KEY` (developer.adzuna.com) and `USAJOBS_API_KEY`\u002F`USAJOBS_EMAIL` (developer.usajobs.gov). Never fabricate listings on a key error.\n\n### How to run a US search\n1. **Specific company** → `greenhouse`\u002F`lever`\u002F`ashby` (no key, instant).\n2. **Federal\u002Fgovernment** → `usajobs`.\n3. **Broad \"X jobs in Y city\"** → `adzuna` (needs key). Without an Adzuna key, fall back to querying a few relevant company boards, and tell the user a key unlocks market-wide search.\n4. Run sources in parallel (separate Bash calls), dedupe by `title`+`company`, present a ranked shortlist (title — company — location — [remote] — salary — url).\n5. Then offer `\u002Fapply \u003Curl>` to tailor a CV + cover letter for any posting.\n\n---\n\n## 🇩🇰 Danish jobs\n\n## Repo location\n\n```\n~\u002Fai-job-search\n```\n\nIf it's missing, clone it: `git clone https:\u002F\u002Fgithub.com\u002FMadsLorentzen\u002Fai-job-search ~\u002Fai-job-search`, then for each of the four CLIs run `bun install` (deps are pinned to `@bunli\u002Fcore@0.7.0` — do NOT bump to \"latest\", 0.8+ pulls a React\u002FOpenTUI runtime that crashes headless).\n\n## CRITICAL run rule — cd into the CLI dir first\n\nBun resolves `node_modules` relative to the **file's** location. Always `cd` into the specific `cli\u002F` dir before running, or bun grabs a wrong global-cache version of `@bunli\u002Fcore` and crashes with `Cannot find module 'react\u002Fjsx-dev-runtime'`.\n\n```bash\ncd ~\u002Fai-job-search\u002F.agents\u002Fskills\u002F\u003Ccli>\u002Fcli && bun run src\u002Fcli.ts \u003Ccommand> [flags]\n```\n\n`\u003Ccli>` ∈ `jobindex-search`, `jobnet-search`, `jobbank-search`, `jobdanmark-search`.\n\nAll commands accept `--format json|table|plain` (default `json`). Pipe JSON to `python3`\u002F`jq` to filter.\n\n---\n\n## The four portals\n\n| CLI dir | Portal | Best for | Live status from this host |\n|---|---|---|---|\n| `jobdanmark-search` | Jobdanmark.dk | Broad market, all sectors | ✅ fully working |\n| `jobnet-search` | Jobnet.dk (govt) | Public-sector, all DK | `suggestions`\u002F`occupations` ✅; `search`\u002F`detail` need a DK\u002Fbrowser session |\n| `jobindex-search` | Jobindex.dk | Largest private board | session\u002Fcookie-gated (204 from datacenter IPs) |\n| `jobbank-search` | Akademikernes Jobbank | Highly-educated \u002F academic | Cloudflare-gated from datacenter IPs |\n\nThe gated ones run, parse, and error-handle cleanly — they return data from the user's Mac (residential IP \u002F logged-in), just not from headless cloud hosts. If a gated CLI returns empty\u002F403\u002F204, say so plainly; don't claim no jobs exist.\n\n---\n\n## Command cheat-sheet\n\n### jobdanmark-search (✅ verified live)\n```bash\ncd ~\u002Fai-job-search\u002F.agents\u002Fskills\u002Fjobdanmark-search\u002Fcli\nbun run src\u002Fcli.ts search --text \"elektriker\" --job-type fuldtid --municipality Odense --limit 10\nbun run src\u002Fcli.ts search --category 227978 --job-type \"fuldtid,deltid\" --page 2\nbun run src\u002Fcli.ts detail \u003Cslug>                 # slug from search results\nbun run src\u002Fcli.ts categories                     # live job counts per category\nbun run src\u002Fcli.ts autocomplete -q \"it\"           # job-title \u002F category IDs\nbun run src\u002Fcli.ts locations -q \"Odense\"          # municipality \u002F zip \u002F region\n```\n\n### jobnet-search\n```bash\ncd ~\u002Fai-job-search\u002F.agents\u002Fskills\u002Fjobnet-search\u002Fcli\nbun run src\u002Fcli.ts suggestions -q \"syge\"                                  # ✅ typeahead titles\nbun run src\u002Fcli.ts occupations --search-string \"sygeplejerske\" --per-page 5  # ✅ ESCO codes\nbun run src\u002Fcli.ts search --search-string \"udvikler\" --region HovedstadenOgBornholm --work-hours FullTime --per-page 10\nbun run src\u002Fcli.ts detail \u003CjobAdId>              # UUID from search\n```\nRegions: `HovedstadenOgBornholm`, `Midtjylland`, `Syddanmark`, `OevrigeSjaelland`, `Nordjylland`. Order: `PublicationDate` | `BestMatch` | `ApplicationDate`.\n\n### jobindex-search\n```bash\ncd ~\u002Fai-job-search\u002F.agents\u002Fskills\u002Fjobindex-search\u002Fcli\nbun run src\u002Fcli.ts search -q \"python\" --jobage 7 --sort date --limit 10   # jobage: 1|7|14|30|9999\nbun run src\u002Fcli.ts detail \u003Cid|url>\n```\nArea filtering isn't reliable via params — put the city in the query: `-q \"python aarhus\"`.\n\n### jobbank-search (academic)\n```bash\ncd ~\u002Fai-job-search\u002F.agents\u002Fskills\u002Fjobbank-search\u002Fcli\nbun run src\u002Fcli.ts search --key \"data scientist\" --location 2 --type 3 --limit 10\nbun run src\u002Fcli.ts detail \u003Cid>\n```\nMulti-value filters are comma-separated: `--type 3,6 --industry 10331`. Code tables live in `jobbank-search\u002Fcli\u002FREADME.md` (type\u002Flocation\u002Fwork-area\u002Feducation\u002Findustry).\n\n---\n\n## How to run a search\n\n1. Resolve the user's intent → pick portal(s). Broad\u002Fany-sector → **jobdanmark** (most reliable). Academic → jobbank. Public-sector → jobnet. Largest private pool → jobindex.\n2. For jobdanmark\u002Fjobnet, use `autocomplete`\u002F`occupations`\u002F`locations` first to turn free text into IDs when filtering precisely.\n3. Run `search` with `--limit` to keep output small; parse the JSON and present a ranked, deduped shortlist (title — company — location — url).\n4. Search several portals in parallel (separate Bash calls) when the user wants broad coverage, then dedupe by title+company across results.\n5. If a portal returns empty\u002Fgated, name which one and why; offer to retry from the logged-in browser.\n\n## Full application workflow (the repo's own commands)\n\nFor end-to-end (fit evaluation → tailored LaTeX CV → cover letter → reviewer pass), the repo ships Claude Code slash commands in `~\u002Fai-job-search\u002F.claude\u002Fcommands\u002F`:\n\n- `\u002Fsetup` — build the candidate profile (fills `CLAUDE.md`, profile files) from `documents\u002F`, a pasted CV, or an interview.\n- `\u002Fscrape` — multi-portal search ranked by fit (uses the CLIs above).\n- `\u002Fapply \u003Curl|text>` — evaluate fit, draft CV (`cv\u002Fmain_\u003Cco>.tex`, compile `lualatex`) + cover letter (`cover_letters\u002F`, compile `xelatex`), reviewer critique, revise.\n\nTo use those, work **inside** `~\u002Fai-job-search` so the repo's `CLAUDE.md` profile + command files are in scope. The profile is currently a blank template — run `\u002Fsetup` there first. Optional salary benchmarking: `salary_lookup.py` + `tools\u002Fconvert_salary_excel.py` (needs a user-supplied `salary_data.json`).\n\n## Quality rules\n\n- Never invent job listings, companies, salaries, or deadlines — every fact comes from CLI JSON output. If gated\u002Fempty, report that, don't fabricate.\n- Keep Danish characters intact (the CLIs decode HTML entities; don't re-mangle æ\u002Fø\u002Få).\n- Don't bump `@bunli\u002Fcore` past 0.7.0 or remove the per-CLI `tsconfig.json` (typecheck OOMs without it).\n","# \u002Fjobsearch — US + Danish Job Search + Application Assistant\n\nThin launcher over the job-portal CLIs in the **ai-job-search** repo. **Default to the US sources** unless the user asks for Denmark. Each CLI emits clean JSON to stdout; errors are JSON on stderr with exit 1. Same run rule everywhere: **`cd` into the CLI's `cli\u002F` dir first** (bun resolves `node_modules` by file location), and never bump `@bunli\u002Fcore` past `0.7.0`.\n\n---\n\n## 🇺🇸 US jobs — `us-jobs` CLI\n\n```bash\ncd ~\u002Fai-job-search\u002F.agents\u002Fskills\u002Fus-jobs\u002Fcli && bun run src\u002Fcli.ts \u003Ccommand> [flags]\n```\n\nFive sources, one normalized output shape: `{meta, results:[{source,id,title,company,location,remote,salary,url,posted}]}`. All accept `--limit` and `--format json|table|plain`.\n\n### Keyless — company job boards (work immediately, ✅ verified live)\nFor \"jobs at \u003Ccompany>\", the **default is `auto`** — it fans out across all three ATSes in parallel so you don't have to know which one the company uses:\n```bash\nbun run src\u002Fcli.ts auto \u003Ccompany> -q engineer --remote --limit 20   # tries greenhouse+lever+ashby, merges, dedupes\n```\n`meta.sources_hit` tells you which ATS actually had the company. `-q` filters by title text. Use the single-source commands only when you already know the ATS:\n```bash\nbun run src\u002Fcli.ts greenhouse \u003Cboard>   --limit 20   # e.g. stripe, airbnb, coinbase\nbun run src\u002Fcli.ts lever      \u003Caccount>  --limit 20   # e.g. palantir\nbun run src\u002Fcli.ts ashby      \u003Cboard>    --remote     # e.g. Notion, Ramp, linear, openai\n```\n`\u003Ccompany>`\u002F`\u003Cboard>`\u002F`\u003Caccount>` is the company's slug in its careers URL (`jobs.lever.co\u002F\u003Caccount>`, `jobs.ashbyhq.com\u002F\u003Cboard>`, `boards.greenhouse.io\u002F\u003Cboard>`). `NOT_FOUND` from `auto` means the company isn't on any of the three keyless ATSes (try Adzuna, or check their careers page for a different ATS).\n\n### Federal — USAJOBS (✅ key configured, verified live)\n```bash\nbun run src\u002Fcli.ts usajobs -q \"data scientist\" --location \"Austin, TX\" --remote --limit 20\n```\n\n### Broad private market — Adzuna (✅ key configured, verified live)\n```bash\nbun run src\u002Fcli.ts adzuna -q \"software engineer\" --where \"Seattle, WA\" --salary-min 120000 --remote --limit 20\n```\n\n**Keys are already set up** (2026-06-04) in `us-jobs\u002Fcli\u002F.env` — bun auto-loads it, so commands just work from the cli dir with no `export` needed. The same keys are also in `~\u002F.zshrc` for interactive terminals. `.env` is gitignored (keys never commit). If you ever see a `MISSING_KEY` error, the `.env` was moved\u002Flost — re-add `ADZUNA_APP_ID`\u002F`ADZUNA_APP_KEY` (developer.adzuna.com) and `USAJOBS_API_KEY`\u002F`USAJOBS_EMAIL` (developer.usajobs.gov). Never fabricate listings on a key error.\n\n### How to run a US search\n1. **Specific company** → `greenhouse`\u002F`lever`\u002F`ashby` (no key, instant).\n2. **Federal\u002Fgovernment** → `usajobs`.\n3. **Broad \"X jobs in Y city\"** → `adzuna` (needs key). Without an Adzuna key, fall back to querying a few relevant company boards, and tell the user a key unlocks market-wide search.\n4. Run sources in parallel (separate Bash calls), dedupe by `title`+`company`, present a ranked shortlist (title — company — location — [remote] — salary — url).\n5. Then offer `\u002Fapply \u003Curl>` to tailor a CV + cover letter for any posting.\n\n---\n\n## 🇩🇰 Danish jobs\n\n## Repo location\n\n```\n~\u002Fai-job-search\n```\n\nIf it's missing, clone it: `git clone https:\u002F\u002Fgithub.com\u002FMadsLorentzen\u002Fai-job-search ~\u002Fai-job-search`, then for each of the four CLIs run `bun install` (deps are pinned to `@bunli\u002Fcore@0.7.0` — do NOT bump to \"latest\", 0.8+ pulls a React\u002FOpenTUI runtime that crashes headless).\n\n## CRITICAL run rule — cd into the CLI dir first\n\nBun resolves `node_modules` relative to the **file's** location. Always `cd` into the specific `cli\u002F` dir before running, or bun grabs a wrong global-cache version of `@bunli\u002Fcore` and crashes with `Cannot find module 'react\u002Fjsx-dev-runtime'`.\n\n```bash\ncd ~\u002Fai-job-search\u002F.agents\u002Fskills\u002F\u003Ccli>\u002Fcli && bun run src\u002Fcli.ts \u003Ccommand> [flags]\n```\n\n`\u003Ccli>` ∈ `jobindex-search`, `jobnet-search`, `jobbank-search`, `jobdanmark-search`.\n\nAll commands accept `--format json|table|plain` (default `json`). Pipe JSON to `python3`\u002F`jq` to filter.\n\n---\n\n## The four portals\n\n| CLI dir | Portal | Best for | Live status from this host |\n|---|---|---|---|\n| `jobdanmark-search` | Jobdanmark.dk | Broad market, all sectors | ✅ fully working |\n| `jobnet-search` | Jobnet.dk (govt) | Public-sector, all DK | `suggestions`\u002F`occupations` ✅; `search`\u002F`detail` need a DK\u002Fbrowser session |\n| `jobindex-search` | Jobindex.dk | Largest private board | session\u002Fcookie-gated (204 from datacenter IPs) |\n| `jobbank-search` | Akademikernes Jobbank | Highly-educated \u002F academic | Cloudflare-gated from datacenter IPs |\n\nThe gated ones run, parse, and error-handle cleanly — they return data from the user's Mac (residential IP \u002F logged-in), just not from headless cloud hosts. If a gated CLI returns empty\u002F403\u002F204, say so plainly; don't claim no jobs exist.\n\n---\n\n## Command cheat-sheet\n\n### jobdanmark-search (✅ verified live)\n```bash\ncd ~\u002Fai-job-search\u002F.agents\u002Fskills\u002Fjobdanmark-search\u002Fcli\nbun run src\u002Fcli.ts search --text \"elektriker\" --job-type fuldtid --municipality Odense --limit 10\nbun run src\u002Fcli.ts search --category 227978 --job-type \"fuldtid,deltid\" --page 2\nbun run src\u002Fcli.ts detail \u003Cslug>                 # slug from search results\nbun run src\u002Fcli.ts categories                     # live job counts per category\nbun run src\u002Fcli.ts autocomplete -q \"it\"           # job-title \u002F category IDs\nbun run src\u002Fcli.ts locations -q \"Odense\"          # municipality \u002F zip \u002F region\n```\n\n### jobnet-search\n```bash\ncd ~\u002Fai-job-search\u002F.agents\u002Fskills\u002Fjobnet-search\u002Fcli\nbun run src\u002Fcli.ts suggestions -q \"syge\"                                  # ✅ typeahead titles\nbun run src\u002Fcli.ts occupations --search-string \"sygeplejerske\" --per-page 5  # ✅ ESCO codes\nbun run src\u002Fcli.ts search --search-string \"udvikler\" --region HovedstadenOgBornholm --work-hours FullTime --per-page 10\nbun run src\u002Fcli.ts detail \u003CjobAdId>              # UUID from search\n```\nRegions: `HovedstadenOgBornholm`, `Midtjylland`, `Syddanmark`, `OevrigeSjaelland`, `Nordjylland`. Order: `PublicationDate` | `BestMatch` | `ApplicationDate`.\n\n### jobindex-search\n```bash\ncd ~\u002Fai-job-search\u002F.agents\u002Fskills\u002Fjobindex-search\u002Fcli\nbun run src\u002Fcli.ts search -q \"python\" --jobage 7 --sort date --limit 10   # jobage: 1|7|14|30|9999\nbun run src\u002Fcli.ts detail \u003Cid|url>\n```\nArea filtering isn't reliable via params — put the city in the query: `-q \"python aarhus\"`.\n\n### jobbank-search (academic)\n```bash\ncd ~\u002Fai-job-search\u002F.agents\u002Fskills\u002Fjobbank-search\u002Fcli\nbun run src\u002Fcli.ts search --key \"data scientist\" --location 2 --type 3 --limit 10\nbun run src\u002Fcli.ts detail \u003Cid>\n```\nMulti-value filters are comma-separated: `--type 3,6 --industry 10331`. Code tables live in `jobbank-search\u002Fcli\u002FREADME.md` (type\u002Flocation\u002Fwork-area\u002Feducation\u002Findustry).\n\n---\n\n## How to run a search\n\n1. Resolve the user's intent → pick portal(s). Broad\u002Fany-sector → **jobdanmark** (most reliable). Academic → jobbank. Public-sector → jobnet. Largest private pool → jobindex.\n2. For jobdanmark\u002Fjobnet, use `autocomplete`\u002F`occupations`\u002F`locations` first to turn free text into IDs when filtering precisely.\n3. Run `search` with `--limit` to keep output small; parse the JSON and present a ranked, deduped shortlist (title — company — location — url).\n4. Search several portals in parallel (separate Bash calls) when the user wants broad coverage, then dedupe by title+company across results.\n5. If a portal returns empty\u002Fgated, name which one and why; offer to retry from the logged-in browser.\n\n## Full application workflow (the repo's own commands)\n\nFor end-to-end (fit evaluation → tailored LaTeX CV → cover letter → reviewer pass), the repo ships Claude Code slash commands in `~\u002Fai-job-search\u002F.claude\u002Fcommands\u002F`:\n\n- `\u002Fsetup` — build the candidate profile (fills `CLAUDE.md`, profile files) from `documents\u002F`, a pasted CV, or an interview.\n- `\u002Fscrape` — multi-portal search ranked by fit (uses the CLIs above).\n- `\u002Fapply \u003Curl|text>` — evaluate fit, draft CV (`cv\u002Fmain_\u003Cco>.tex`, compile `lualatex`) + cover letter (`cover_letters\u002F`, compile `xelatex`), reviewer critique, revise.\n\nTo use those, work **inside** `~\u002Fai-job-search` so the repo's `CLAUDE.md` profile + command files are in scope. The profile is currently a blank template — run `\u002Fsetup` there first. Optional salary benchmarking: `salary_lookup.py` + `tools\u002Fconvert_salary_excel.py` (needs a user-supplied `salary_data.json`).\n\n## Quality rules\n\n- Never invent job listings, companies, salaries, or deadlines — every fact comes from CLI JSON output. If gated\u002Fempty, report that, don't fabricate.\n- Keep Danish characters intact (the CLIs decode HTML entities; don't re-mangle æ\u002Fø\u002Få).\n- Don't bump `@bunli\u002Fcore` past 0.7.0 or remove the per-CLI `tsconfig.json` (typecheck OOMs without it).\n",{"name":5,"description":7},"main","e7b39d448c7d905feed2b044d3ccdf60822596ab","skills\u002Fjobsearch",0,"https:\u002F\u002Fgithub.com\u002Fbarkleesanders\u002Fclaude-code-starter","https:\u002F\u002Fgithub.com\u002Fbarkleesanders","barkleesanders.com","Stay Hungry, Stay Foolish","2026-11-06 08:47:06",{"source":8,"skillName":5,"commands":34,"prompt":44,"zipAvailable":45},[35,38,41],{"tool":36,"command":37},"npx","npx skills add barkleesanders\u002Fclaude-code-starter --skill jobsearch",{"tool":39,"command":40},"bunx","bunx skills add barkleesanders\u002Fclaude-code-starter --skill jobsearch",{"tool":42,"command":43},"pnpm","pnpm dlx skills add barkleesanders\u002Fclaude-code-starter --skill jobsearch","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fbarkleesanders\u002Fclaude-code-starter\u002Fblob\u002Fmain\u002Fskills\u002Fjobsearch\u002FSKILL.md",true,[],[48,54,61,68,75,82],{"skillNo":49,"name":50,"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":51,"categories":52,"keywords":53,"publishedAt":19},"01M7AWV75B37B6YKG7ZKGZ783F","zi-search","https:\u002F\u002Fgithub.com\u002Fbarkleesanders\u002Fclaude-code-starter\u002Fblob\u002Fmain\u002Fskills\u002Fzi-search\u002FSKILL.md",[],[],{"skillNo":55,"name":56,"nameZh":6,"description":57,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":58,"categories":59,"keywords":60,"publishedAt":19},"01M7AWTZA3JGK1T7V1HS543FG7","visualise","Render inline interactive visuals — SVG diagrams, HTML widgets, charts, and explainers — directly in the conversation. Use this skill whenever the user asks to visualize, diagram, chart, illustrate, or explain something visually, or when an explanation would genuinely benefit from a spatial\u002Finteractive diagram rather than text. Also triggers for: flowcharts, architecture diagrams, data visualizations, interactive explainers, comparison layouts, UI mockups, and any request containing 'show me', 'draw', 'map out', 'visualize', or 'diagram'. Even when the user doesn't explicitly ask for a visual, use this skill proactively when the topic has spatial, sequential, or systemic relationships that a diagram would clarify better than prose.","https:\u002F\u002Fgithub.com\u002Fbarkleesanders\u002Fclaude-code-starter\u002Fblob\u002Fmain\u002Fskills\u002Fvisualise\u002FSKILL.md",[],[],{"skillNo":62,"name":63,"nameZh":6,"description":64,"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},"01M7AWTR056C3P9R3VS8YVM2WX","video","When the user wants to create, generate, or produce video content using AI tools or programmatic frameworks. Also use when the user mentions 'video production,' 'AI video,' 'Remotion,' 'Hyperframes,' 'HeyGen,' 'Synthesia,' 'Veo,' 'Runway,' 'Kling,' 'Pika,' 'video generation,' 'AI avatar,' 'talking head video,' 'programmatic video,' 'video template,' 'explainer video,' 'product demo video,' 'video pipeline,' or 'make me a video.' Use this for video creation, generation, and production workflows. For video content strategy and what to post, see social-content. For paid video ad creative, see ad-creative.","https:\u002F\u002Fgithub.com\u002Fbarkleesanders\u002Fclaude-code-starter\u002Fblob\u002Fmain\u002Fskills\u002Fvideo\u002FSKILL.md",[],[],{"skillNo":69,"name":70,"nameZh":6,"description":71,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":72,"categories":73,"keywords":74,"publishedAt":19},"01M7AWTHTWBMTYZPMHHEYX3TTG","manim-video","Production pipeline for mathematical and technical animations using Manim Community Edition. Creates 3Blue1Brown-style explainer videos, algorithm visualizations, equation derivations, architecture diagrams, and data stories. Use when users request: animated explanations, math animations, concept visualizations, algorithm walkthroughs, technical explainers, 3Blue1Brown style videos, or any programmatic animation with geometric\u002Fmathematical content.","https:\u002F\u002Fgithub.com\u002Fbarkleesanders\u002Fclaude-code-starter\u002Fblob\u002Fmain\u002Fskills\u002Fvideo-use\u002Fskills\u002Fmanim-video\u002FSKILL.md",[],[],{"skillNo":76,"name":77,"nameZh":6,"description":78,"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},"01M7AWTCXDJ9NWFQW51ENM8A2Z","video-use","Edit any video by conversation. Transcribe, cut, color grade, generate overlay animations, burn subtitles — for talking heads, montages, tutorials, travel, interviews. No presets, no menus. Ask questions, confirm the plan, execute, iterate, persist. Production-correctness rules are hard; everything else is artistic freedom.","https:\u002F\u002Fgithub.com\u002Fbarkleesanders\u002Fclaude-code-starter\u002Fblob\u002Fmain\u002Fskills\u002Fvideo-use\u002FSKILL.md",[],[],{"skillNo":83,"name":84,"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":85,"categories":86,"keywords":87,"publishedAt":19},"01M7AWTA1D5FP895S33GHSA2G5","ui-improver","https:\u002F\u002Fgithub.com\u002Fbarkleesanders\u002Fclaude-code-starter\u002Fblob\u002Fmain\u002Fskills\u002Fui-improver\u002FSKILL.md",[],[]]