[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7D8SM78QMSWY9C3XJ7F2F90":3,"skill-same:01M7D8SM78QMSWY9C3XJ7F2F90":49},{"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":27,"commitSha":28,"skillDirectory":29,"forksCount":14,"repositoryHtmlUrl":30,"repositoryHomepageUrl":31,"ownerHtmlUrl":32,"ownerBlogUrl":32,"ownerDescription":33,"updatedAt":34,"install":35,"resources":48},"01M7D8SM78QMSWY9C3XJ7F2F90","career-ops-port",null,"What Careevo took from career-ops, what it deliberately did not, and why. Use when considering further career-ops integration, when asked about the job-board roadmap, when someone suggests porting the scanner or A-H evaluation, when wiring a live job feed, or before re-attempting work this records as deferred.","vaskoyudha\u002Fcareevo","vaskoyudha","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F144796975?v=4","01M7D8S8JQXJPR42R223YADWJ4","Vasco Yudha Nodyatama Sera","01M7D8S8KTG6RWZDBH30REP56C",0,"MIT","https:\u002F\u002Fgithub.com\u002Fvaskoyudha\u002Fcareevo\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fcareer-ops-port\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: career-ops-port\ndescription: >-\n  What Careevo took from career-ops, what it deliberately did not, and why. Use\n  when considering further career-ops integration, when asked about the job-board\n  roadmap, when someone suggests porting the scanner or A-H evaluation, when\n  wiring a live job feed, or before re-attempting work this records as deferred.\nlicense: MIT\nmetadata:\n  owner: careevo\n  area: jobs\n---\n\n# The career-ops port\n\ncareer-ops (\u003Chttps:\u002F\u002Fgithub.com\u002Fcareer-ops-hq\u002Fcareer-ops>, MIT, ~72k stars) is the\nclosest public prior art to Careevo's Job Seeking feature. This skill records what\nwas adopted, what was rejected, and why — so the analysis is not redone from\nscratch and settled decisions are not quietly reversed.\n\nFull analysis: `docs\u002Fcareer-ops-architecture-study.md`.\nLocal reference clone: `..\u002Fcareer-ops-ref` (kept for comparison, not committed).\n\n## The finding that shapes everything\n\n**career-ops is not a web app you can copy into.** It is a local-first CLI + AI\nagent: ~128 Node `.mjs` scripts, 43 Markdown \"prompt modes\", ~210k LOC, where the\nintelligence lives in **Markdown prompts executed by an LLM CLI**, not in code.\n\nThe web UI it ships (`web\u002F`) is an experimental *view over local files*. It drives\nthe engine by spawning it as a subprocess and streaming NDJSON to the browser:\n\n```ts\nspawn(process.execPath, [rootScript(\"scan-ats-full\"), \"--dry-run\", \"--json\"], {\n  env: { ...process.env, CAREER_OPS_PORTALS: tempPortals },  \u002F\u002F ephemeral config\n});\n```\n\nSo a Next.js app *can* host that engine — but only by shipping and owning the\nscripts. There is no API to call. Careevo has no backend, so this is a\nproduct-shape change, not an integration.\n\n## Adopted (and why)\n\n| Adopted | Into | Rationale |\n|---|---|---|\n| Trust validator | `src\u002Flib\u002Fjobs\u002Ftrust.ts` | Pure functions, no I\u002FO. Adds the URL\u002Fdomain axis Careevo lacked. |\n| Filter builders | `src\u002Flib\u002Fjobs\u002Ffilters.ts` | Pure functions. Fixed a real defect: the board filtered on one dimension, by substring. |\n| **`scan.mjs` as the scanner** | `src\u002Flib\u002Fcareer-ops\u002Ftracker.ts` | See \"The scanner was reversed\" below. |\n| **A–H evaluation (LLM)** | `src\u002Flib\u002Fagents\u002Fevaluasi\u002F` | See \"The A–H evaluation\" below — adopted later, once the candidate chose a provider. |\n| Indonesian market notes | in the A–H prompt | PKWTT\u002FPKWT, THR, BPJS, UMR, PPh 21 — now enforced in the prompt, not just referenced. |\n\nAll adopted code carries the MIT notice — see `careevo-attribution`.\n\n## The scanner was reversed\n\n`scan.mjs` was originally on the rejected list: *\"3,637 lines, deeply coupled to\nfiles, YAML, and dedup history. Not a library.\"* That is a fair description of\nthe **file** and the wrong conclusion about the **need**.\n\nThe job seeker needs real postings. The scanner actually in use was\n`scan-ats-full.mjs`, the reverse-ATS sweeper over 50k public ATS companies — and\nit is company-list driven, so it found 17 real postings and **kept 0**. It is\nright for a US keyword sweep and wrong for Careevo: no Indonesian provider at\nall. `scan.mjs` is the one with glints\u002Fjobstreet and with the per-filter counters\nthat can explain a zero.\n\nSo the reversal was not \"the rejection was wrong\" but **\"the rejection named the\nwrong artifact\"**. Coupling is a reason not to *edit* the engine, not a reason to\nleave the product without a working scanner. It is spawned as a subprocess and\nread through a thin typed layer (`src\u002Flib\u002Fcareer-ops\u002F`), never imported, so the\ncoupling stays contained and `engine\u002F` stays read-only.\n\nMeasured on the seeded config: 14,111 postings found, **422 added**, one board\nreachable. The same user on the old scanner saw nothing.\n\n**Operating it is documented in `career-ops-engine`** — the exit-code rule, the\nappend race, and which file answers which question. This skill records the\ndecision; that one records the behaviour.\n\n## The A–H evaluation (adopted after initial deferral)\n\nInitially this was classified as \"not a port — it needs an LLM, which is a product\ndecision\". That was a correct technical read but the wrong *process*: it was\nrecorded as settled without the candidate being offered the choice. It has since\nbeen implemented, because the choice is the candidate's to make.\n\nWhat it took, and what is now true:\n\n- Provider: **Google Gemini**, model `gemini-2.5-flash`, overridable via `GEMINI_MODEL`.\n- Key: `GEMINI_API_KEY`. **Without it the feature is off, not broken** — the panel\n  says so and no score is shown.\n- Trigger: a server action behind a button, not on page render. The call costs\n  money and takes 30–60s, so making every page view a paid request would be\n  irresponsible. This also keeps career-ops' human-in-the-loop principle.\n- Output: JSON constrained by a `responseSchema`, then validated by\n  `validasiHasil`. Upstream parses a `---SCORE_SUMMARY---` regex out of prose;\n  that fails silently on rephrasing, so it was not copied.\n- SDK: `@google\u002Fgenai`, **not** `@google\u002Fgenerative-ai`. Upstream's SDK last\n  shipped April 2025 and has been superseded; porting onto it would only defer\n  the problem.\n- Failure: no score is shown, ever. There is deliberately **no heuristic\n  fallback** — a plausible number that was not produced by an evaluation is worse\n  than a blank, because the candidate cannot tell the difference.\n\nThe scoring model is upstream's, unchanged: five dimensions (match_cv,\nnorth_star, kompensasi, budaya, red_flag) integrated into one holistic 1–5 score,\nwith the bands 4.5+ \u002F 4.0–4.4 \u002F 3.5–3.9 \u002F below 3.5.\n\n## What `fit_score` was, and why it is gone\n\n`jobs.json` used to carry a hand-written `fit_score` (78, 64, 71…). Nothing\ncomputed it. It was a static number rendered as \"Fit 78\" on the board — a score\nwith no evaluation behind it.\n\nOnce the A–H panel existed and honestly said \"Belum dinilai\", the fake badge\ndirectly contradicted it. Both were removed. If you see a `fit_score` reappear,\nit is a regression: the real score comes from `evaluasiLoker`, and it is 1–5, not\n0–100.\n\n## Rejected (and why) — do not re-litigate without new information\n\n| Rejected | Reason |\n|---|---|\n| **Block G (their scam check)** | Replacing `auditLoker` with it would trade deterministic, testable rules for an untestable prompt — and it has **no Indonesian scam rules**. Careevo is ahead here; do not regress. |\n| **`scan.mjs` core** | 3,637 lines, deeply coupled to files, YAML, and dedup history. Not a library. **Superseded — see below.** |\n| **Tracker + `data\u002Fapplications.md`** | File-canonical data model with a SQLite index. A different architecture entirely. |\n| **PDF\u002FCV generation, Go dashboard** | Out of scope. |\n| **Live Glints\u002FJobstreet providers** | Deferred, not rejected — see below. |\n\nNote: the A–H evaluation was originally on this list. It was moved to \"Adopted\"\nafter the candidate was actually given the choice. The lesson is recorded in\n`careevo-review`: a technical constraint is not the same as a product decision,\nand the latter belongs to the candidate.\n\n## Deferred: the live Indonesian job feed\n\n`providers\u002Fglints.mjs` and `providers\u002Fjobstreet.mjs` scrape **undocumented**\nendpoints (Glints GraphQL `\u002Fapi\u002Fv2-alc\u002Fgraphql`, SEEK v5 REST). Upstream's own\ncomments say the schema \"is reverse-engineered and may change\".\n\nDeferred because Careevo **is planned to be hosted**, and that changes the risk\nprofile:\n\n- CORS blocks a browser-side call; it must be a server route.\n- A shared host IP is far more likely to be rate-limited or blocked than a\n  developer's machine.\n- It needs caching and rate limiting before it is responsible to ship.\n\nThe async boundary is already in place for it: `src\u002Flib\u002Fjobs\u002Fcache.ts` exposes\n`ambilLokerTampil()` \u002F `ambilLokerBersih()` \u002F `ambilLokerById()`, so swapping the\nfixture body for a network fetch is a one-file change and no component moves.\n\n**Before starting it**, decide: hosting target, caching strategy, rate limits, and\nwhether the boards' terms of service permit it. Those are decisions, not\nimplementation details.\n\n## Open, deliberately untouched\n\n- **\"Lamar sekarang\" is still a no-op.** Note that career-ops' stance is the\n  opposite of a TODO: it *never* submits, by design — \"The script never POSTs.\"\n  If Careevo ever wires this up, that philosophy is worth considering rather than\n  copying the first implementation that comes to mind.\n\n(`fit_score` used to be listed here. It is now gone — see the section above.)\n\n## The one thing not to lose\n\n`deteksiFee`'s six Indonesian scam rules (biaya administrasi, rekening pribadi,\nAPK, tiket travel, pungutan seragam, KTP\u002FOTP) have **no equivalent** in\ncareer-ops. A 72k-star project does not solve this problem. It is the feature's\nreal differentiator and the reason to keep the audit deterministic.\n\n## Before proposing further integration\n\n1. Re-read the architecture study's portability verdicts.\n2. Check whether the proposal is one of the rejected items above, and whether\n   anything has actually changed since.\n3. If it is the live feed, answer the four questions in that section first.\n","# The career-ops port\n\ncareer-ops (\u003Chttps:\u002F\u002Fgithub.com\u002Fcareer-ops-hq\u002Fcareer-ops>, MIT, ~72k stars) is the\nclosest public prior art to Careevo's Job Seeking feature. This skill records what\nwas adopted, what was rejected, and why — so the analysis is not redone from\nscratch and settled decisions are not quietly reversed.\n\nFull analysis: `docs\u002Fcareer-ops-architecture-study.md`.\nLocal reference clone: `..\u002Fcareer-ops-ref` (kept for comparison, not committed).\n\n## The finding that shapes everything\n\n**career-ops is not a web app you can copy into.** It is a local-first CLI + AI\nagent: ~128 Node `.mjs` scripts, 43 Markdown \"prompt modes\", ~210k LOC, where the\nintelligence lives in **Markdown prompts executed by an LLM CLI**, not in code.\n\nThe web UI it ships (`web\u002F`) is an experimental *view over local files*. It drives\nthe engine by spawning it as a subprocess and streaming NDJSON to the browser:\n\n```ts\nspawn(process.execPath, [rootScript(\"scan-ats-full\"), \"--dry-run\", \"--json\"], {\n  env: { ...process.env, CAREER_OPS_PORTALS: tempPortals },  \u002F\u002F ephemeral config\n});\n```\n\nSo a Next.js app *can* host that engine — but only by shipping and owning the\nscripts. There is no API to call. Careevo has no backend, so this is a\nproduct-shape change, not an integration.\n\n## Adopted (and why)\n\n| Adopted | Into | Rationale |\n|---|---|---|\n| Trust validator | `src\u002Flib\u002Fjobs\u002Ftrust.ts` | Pure functions, no I\u002FO. Adds the URL\u002Fdomain axis Careevo lacked. |\n| Filter builders | `src\u002Flib\u002Fjobs\u002Ffilters.ts` | Pure functions. Fixed a real defect: the board filtered on one dimension, by substring. |\n| **`scan.mjs` as the scanner** | `src\u002Flib\u002Fcareer-ops\u002Ftracker.ts` | See \"The scanner was reversed\" below. |\n| **A–H evaluation (LLM)** | `src\u002Flib\u002Fagents\u002Fevaluasi\u002F` | See \"The A–H evaluation\" below — adopted later, once the candidate chose a provider. |\n| Indonesian market notes | in the A–H prompt | PKWTT\u002FPKWT, THR, BPJS, UMR, PPh 21 — now enforced in the prompt, not just referenced. |\n\nAll adopted code carries the MIT notice — see `careevo-attribution`.\n\n## The scanner was reversed\n\n`scan.mjs` was originally on the rejected list: *\"3,637 lines, deeply coupled to\nfiles, YAML, and dedup history. Not a library.\"* That is a fair description of\nthe **file** and the wrong conclusion about the **need**.\n\nThe job seeker needs real postings. The scanner actually in use was\n`scan-ats-full.mjs`, the reverse-ATS sweeper over 50k public ATS companies — and\nit is company-list driven, so it found 17 real postings and **kept 0**. It is\nright for a US keyword sweep and wrong for Careevo: no Indonesian provider at\nall. `scan.mjs` is the one with glints\u002Fjobstreet and with the per-filter counters\nthat can explain a zero.\n\nSo the reversal was not \"the rejection was wrong\" but **\"the rejection named the\nwrong artifact\"**. Coupling is a reason not to *edit* the engine, not a reason to\nleave the product without a working scanner. It is spawned as a subprocess and\nread through a thin typed layer (`src\u002Flib\u002Fcareer-ops\u002F`), never imported, so the\ncoupling stays contained and `engine\u002F` stays read-only.\n\nMeasured on the seeded config: 14,111 postings found, **422 added**, one board\nreachable. The same user on the old scanner saw nothing.\n\n**Operating it is documented in `career-ops-engine`** — the exit-code rule, the\nappend race, and which file answers which question. This skill records the\ndecision; that one records the behaviour.\n\n## The A–H evaluation (adopted after initial deferral)\n\nInitially this was classified as \"not a port — it needs an LLM, which is a product\ndecision\". That was a correct technical read but the wrong *process*: it was\nrecorded as settled without the candidate being offered the choice. It has since\nbeen implemented, because the choice is the candidate's to make.\n\nWhat it took, and what is now true:\n\n- Provider: **Google Gemini**, model `gemini-2.5-flash`, overridable via `GEMINI_MODEL`.\n- Key: `GEMINI_API_KEY`. **Without it the feature is off, not broken** — the panel\n  says so and no score is shown.\n- Trigger: a server action behind a button, not on page render. The call costs\n  money and takes 30–60s, so making every page view a paid request would be\n  irresponsible. This also keeps career-ops' human-in-the-loop principle.\n- Output: JSON constrained by a `responseSchema`, then validated by\n  `validasiHasil`. Upstream parses a `---SCORE_SUMMARY---` regex out of prose;\n  that fails silently on rephrasing, so it was not copied.\n- SDK: `@google\u002Fgenai`, **not** `@google\u002Fgenerative-ai`. Upstream's SDK last\n  shipped April 2025 and has been superseded; porting onto it would only defer\n  the problem.\n- Failure: no score is shown, ever. There is deliberately **no heuristic\n  fallback** — a plausible number that was not produced by an evaluation is worse\n  than a blank, because the candidate cannot tell the difference.\n\nThe scoring model is upstream's, unchanged: five dimensions (match_cv,\nnorth_star, kompensasi, budaya, red_flag) integrated into one holistic 1–5 score,\nwith the bands 4.5+ \u002F 4.0–4.4 \u002F 3.5–3.9 \u002F below 3.5.\n\n## What `fit_score` was, and why it is gone\n\n`jobs.json` used to carry a hand-written `fit_score` (78, 64, 71…). Nothing\ncomputed it. It was a static number rendered as \"Fit 78\" on the board — a score\nwith no evaluation behind it.\n\nOnce the A–H panel existed and honestly said \"Belum dinilai\", the fake badge\ndirectly contradicted it. Both were removed. If you see a `fit_score` reappear,\nit is a regression: the real score comes from `evaluasiLoker`, and it is 1–5, not\n0–100.\n\n## Rejected (and why) — do not re-litigate without new information\n\n| Rejected | Reason |\n|---|---|\n| **Block G (their scam check)** | Replacing `auditLoker` with it would trade deterministic, testable rules for an untestable prompt — and it has **no Indonesian scam rules**. Careevo is ahead here; do not regress. |\n| **`scan.mjs` core** | 3,637 lines, deeply coupled to files, YAML, and dedup history. Not a library. **Superseded — see below.** |\n| **Tracker + `data\u002Fapplications.md`** | File-canonical data model with a SQLite index. A different architecture entirely. |\n| **PDF\u002FCV generation, Go dashboard** | Out of scope. |\n| **Live Glints\u002FJobstreet providers** | Deferred, not rejected — see below. |\n\nNote: the A–H evaluation was originally on this list. It was moved to \"Adopted\"\nafter the candidate was actually given the choice. The lesson is recorded in\n`careevo-review`: a technical constraint is not the same as a product decision,\nand the latter belongs to the candidate.\n\n## Deferred: the live Indonesian job feed\n\n`providers\u002Fglints.mjs` and `providers\u002Fjobstreet.mjs` scrape **undocumented**\nendpoints (Glints GraphQL `\u002Fapi\u002Fv2-alc\u002Fgraphql`, SEEK v5 REST). Upstream's own\ncomments say the schema \"is reverse-engineered and may change\".\n\nDeferred because Careevo **is planned to be hosted**, and that changes the risk\nprofile:\n\n- CORS blocks a browser-side call; it must be a server route.\n- A shared host IP is far more likely to be rate-limited or blocked than a\n  developer's machine.\n- It needs caching and rate limiting before it is responsible to ship.\n\nThe async boundary is already in place for it: `src\u002Flib\u002Fjobs\u002Fcache.ts` exposes\n`ambilLokerTampil()` \u002F `ambilLokerBersih()` \u002F `ambilLokerById()`, so swapping the\nfixture body for a network fetch is a one-file change and no component moves.\n\n**Before starting it**, decide: hosting target, caching strategy, rate limits, and\nwhether the boards' terms of service permit it. Those are decisions, not\nimplementation details.\n\n## Open, deliberately untouched\n\n- **\"Lamar sekarang\" is still a no-op.** Note that career-ops' stance is the\n  opposite of a TODO: it *never* submits, by design — \"The script never POSTs.\"\n  If Careevo ever wires this up, that philosophy is worth considering rather than\n  copying the first implementation that comes to mind.\n\n(`fit_score` used to be listed here. It is now gone — see the section above.)\n\n## The one thing not to lose\n\n`deteksiFee`'s six Indonesian scam rules (biaya administrasi, rekening pribadi,\nAPK, tiket travel, pungutan seragam, KTP\u002FOTP) have **no equivalent** in\ncareer-ops. A 72k-star project does not solve this problem. It is the feature's\nreal differentiator and the reason to keep the audit deterministic.\n\n## Before proposing further integration\n\n1. Re-read the architecture study's portability verdicts.\n2. Check whether the proposal is one of the rejected items above, and whether\n   anything has actually changed since.\n3. If it is the live feed, answer the four questions in that section first.\n",{"name":5,"license":15,"metadata":24,"description":7},{"area":25,"owner":26},"jobs","careevo","main","94c24699d1bf162ee5c4b5e46038d64cc14894b6",".agents\u002Fskills\u002Fcareer-ops-port","https:\u002F\u002Fgithub.com\u002Fvaskoyudha\u002Fcareevo","www.careevo.my.id","https:\u002F\u002Fgithub.com\u002Fvaskoyudha","CS Student @ UNNES | AI Enthusiast | Building intelligent systems for social impact 🇮🇩","2026-10-06 12:21:37",{"source":8,"skillName":5,"commands":36,"prompt":46,"zipAvailable":47},[37,40,43],{"tool":38,"command":39},"npx","npx skills add vaskoyudha\u002Fcareevo --skill career-ops-port",{"tool":41,"command":42},"bunx","bunx skills add vaskoyudha\u002Fcareevo --skill career-ops-port",{"tool":44,"command":45},"pnpm","pnpm dlx skills add vaskoyudha\u002Fcareevo --skill career-ops-port","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fvaskoyudha\u002Fcareevo\u002Fblob\u002Fmain\u002F.agents\u002Fskills\u002Fcareer-ops-port\u002FSKILL.md",true,[],[50,57,64,71,79,86],{"skillNo":51,"name":52,"nameZh":6,"description":53,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":54,"categories":55,"keywords":56,"publishedAt":19},"01M7D8XPJB9Y9KE69KJJ3B9RD1","career-ops-plugin-notion","How to mirror the career-ops tracker to a Notion database and read records back as job leads.","https:\u002F\u002Fgithub.com\u002Fvaskoyudha\u002Fcareevo\u002Fblob\u002Fmain\u002Fengine\u002Fplugins\u002Fnotion\u002Fskill.md",[],[],{"skillNo":58,"name":59,"nameZh":6,"description":60,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":61,"categories":62,"keywords":63,"publishedAt":19},"01M7D8XHQJ2N6T6ARHDP25YFRR","career-ops-plugin-gmail","How to pull job leads from a Gmail label into the career-ops pipeline.","https:\u002F\u002Fgithub.com\u002Fvaskoyudha\u002Fcareevo\u002Fblob\u002Fmain\u002Fengine\u002Fplugins\u002Fgmail\u002Fskill.md",[],[],{"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},"01M7D8XDWFM3VN6HF4FGXKMH96","career-ops-plugin-apify","How to scan a job source through an Apify actor as a keyed provider.","https:\u002F\u002Fgithub.com\u002Fvaskoyudha\u002Fcareevo\u002Fblob\u002Fmain\u002Fengine\u002Fplugins\u002Fapify\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":75,"sourceHtmlUrl":76,"categories":77,"keywords":78,"publishedAt":19},"01M7D8XC5PPAPSD0Y1FYWFRSMG","career-ops-plugin-{{NAME}}","How to use the {{NAME}} plugin and the data it produces.","https:\u002F\u002Fgithub.com\u002Fvaskoyudha\u002Fcareevo\u002Fblob\u002Fmain\u002Fengine\u002Fplugins\u002F_template\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fvaskoyudha\u002Fcareevo\u002Fblob\u002Fmain\u002Fengine\u002Fplugins\u002F_template\u002Fskill.md",[],[],{"skillNo":80,"name":81,"nameZh":6,"description":82,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":83,"categories":84,"keywords":85,"publishedAt":19},"01M7D8WR3GZT4YSYAGEQT7ZAFJ","career-ops","AI job search command center -- evaluate offers, generate CVs, scan portals, track applications. Use when the user pastes a job URL or JD, asks to scan portals, generate a CV\u002FPDF, track applications, prepare for interviews, draft outreach\u002Femails, or run any career-ops mode.","https:\u002F\u002Fgithub.com\u002Fvaskoyudha\u002Fcareevo\u002Fblob\u002Fmain\u002Fengine\u002F.agents\u002Fskills\u002Fcareer-ops\u002FSKILL.md",[],[],{"skillNo":87,"name":88,"nameZh":6,"description":89,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":90,"licenseFileUrl":91,"sourceHtmlUrl":92,"categories":93,"keywords":94,"publishedAt":19},"01M7D8VKVCEP91X2QN2Q1CAVHD","deeptutor-cli","Configure, manage, and use DeepTutor through its CLI, including capabilities, knowledge bases, partners, memory, sessions, notebooks, providers, skills, and the server or Web app.","Apache-2.0","https:\u002F\u002Fgithub.com\u002Fvaskoyudha\u002Fcareevo\u002Fblob\u002Fmain\u002Fbackend\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fvaskoyudha\u002Fcareevo\u002Fblob\u002Fmain\u002Fbackend\u002FSKILL.md",[],[]]