linkedin-relay-mcp 3.10.0 → 3.11.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
@@ -12,7 +12,8 @@ in the request path.
12
12
  ## Three things to know before you call anything
13
13
 
14
14
  1. **Every call spends a budget that cannot be topped up.** Caps are deliberately low (30 profile
15
- reads a day, 25 searches, 250 requests total) because LinkedIn restricts accounts for *read*
15
+ reads a day, 25 searches, 350 requests total; job hunting has its own 150 job searches and 60
16
+ job reads) because LinkedIn restricts accounts for *read*
16
17
  velocity alone — several documented restrictions involved no automation and no writes, just fast
17
18
  browsing. Check `meta.budget.remaining`; when it's low, tell the user and stop rather than
18
19
  spending the rest.
@@ -58,7 +59,7 @@ Out-of-network people come back as `name: "LinkedIn Member"` with a `/headless`
58
59
  LinkedIn behaviour, not a bug — LinkedIn will not name them to this account. Say so rather than
59
60
  presenting it as a data problem.
60
61
 
61
- ### `jobs` — 1 search call per 25 results. Job search with every filter.
62
+ ### `jobs` — 1 job-search call per 100 results. Job search with every filter.
62
63
  ```
63
64
  jobs keywords:"<terms>" [location:"<place>" | geoId:"<id>"] [distance:<miles>]
64
65
  [posted:"24h|week|month|<N>h|<N>d"] [experience:[internship|entry|associate|mid-senior|director|executive]]
@@ -70,28 +71,32 @@ jobs keywords:"<terms>" [location:"<place>" | geoId:"<id>"] [distance:<miles>]
70
71
  Each item: `id`, `title`, `company`, `location`, `workplace`, `listed` (YYYY-MM-DD), `easyApply`,
71
72
  `promoted`, `salary` (when the card shows one), `insights`, `url`. `compact` gives one line per job.
72
73
 
73
- - **Rank on the rows, then read the finalists with `job`.** `details` reads *every* row (1 profile
74
- call each, 30 a day) to add the apply link, applicants and employment type.
74
+ - **Rank on the rows, then read the finalists with `job`.** `details` reads every row in full, 100
75
+ per job read, to add the apply link, applicants and employment type.
75
76
  - **Check `data.location`.** A typed location is resolved by LinkedIn, and the place it chose is
76
77
  reported there. "Vienna, Austria" resolved to one district of Vienna. Pass `geoId` for an exact
77
78
  place; ids are in `job_filters` under `populatedPlace`.
78
79
  - **Paging:** `meta.claimedCount` is LinkedIn's total, `meta.next` is the `start` for the next
79
80
  batch, and `meta.state` is `complete` only when the results ran out. LinkedIn serves at most 1000
80
81
  per search; narrow the query rather than paging deep.
81
- - For several searches (keyword variants), keep each one narrow with filters. Every 25 rows costs a
82
- search call, and the day has 25.
82
+ - Job hunting has its own budget: 150 job-search calls and 60 job reads a day, apart from the
83
+ people and company `search`. A search call returns 100 jobs, so ten keyword searches of a few
84
+ hundred results each take about 30 calls. The same job turns
85
+ up under several keywords; merge the lists by `id` before reading any in full.
83
86
 
84
- ### `job` — 1 profile call. One job, in full.
87
+ ### `job` — 1 job read per 100 jobs. Jobs, in full.
85
88
  ```
86
- job job:"<id | /jobs/view/ URL | ?currentJobId= URL>"
89
+ job job:"<id | /jobs/view/ URL | ?currentJobId= URL>" or job job:["<id>", "<id>", …]
87
90
  ```
91
+ One id returns the job; a list returns `{items, meta}`, read 100 per call, so pass all your finalists
92
+ at once rather than one call each. `meta.unread` lists ids LinkedIn did not return.
88
93
  `description`, `apply: {type: "external" | "easy-apply", url}`, `company`, `companyUrl`,
89
94
  `location`, `workplace`, `employmentType`, `experienceLevel`, `industries`, `functions`, `listed`,
90
95
  `expires`, `state` (`LISTED` or `CLOSED`), `applicants`, `ats` (the employer's applicant tracking
91
96
  system), and `applied` / `saved` for this user. For an external application, `apply.url` goes to
92
97
  the employer's own site.
93
98
 
94
- ### `job-filters` — 1 search call. The knobs, with ids.
99
+ ### `job-filters` — 1 job-search call. The knobs, with ids.
95
100
  ```
96
101
  job_filters keywords:"<terms>" [location | geoId] [any jobs filter]
97
102
  ```
package/dist/cli.js CHANGED
@@ -946,6 +946,21 @@ var CAPS = {
946
946
  },
947
947
  search: { perDay: 25, provenance: "guessed", note: "no corroborated figure exists" },
948
948
  page: { perDay: 60, provenance: "guessed", note: "thread/feed pagination" },
949
+ // Set on 2026-10-14 by the owner's decision, for job hunting: about ten
950
+ // keyword searches read in full (25 jobs a page, often 100-300 a search),
951
+ // then the finalists read for their apply links. Kept apart from `search`
952
+ // so a hunt neither uses up nor is stopped by people and company research.
953
+ // Guesses like every other cap; the client's 3-15s gap still paces each call.
954
+ jobSearch: {
955
+ perDay: 150,
956
+ provenance: "guessed",
957
+ note: "job search pages and filter lists; set for ~10 keyword searches a day"
958
+ },
959
+ jobRead: {
960
+ perDay: 60,
961
+ provenance: "guessed",
962
+ note: "one job read in full, for its apply link"
963
+ },
949
964
  write: {
950
965
  // Raised from 10 on 2026-08-14 by the owner's explicit decision, after 10
951
966
  // blocked ordinary development on a tool whose writes are the point — a
@@ -961,7 +976,9 @@ var CAPS = {
961
976
  provenance: "guessed",
962
977
  note: "raised from 10 for development; still below the ~20/day connection lore per action type"
963
978
  },
964
- global: { perDay: 250, provenance: "guessed", note: "all classes combined" }
979
+ // Raised from 250 with the job classes, so a full hunt (~120 searches and
980
+ // 60 reads) still leaves room for everything else.
981
+ global: { perDay: 350, provenance: "guessed", note: "all classes combined" }
965
982
  };
966
983
  var MIN_GAP_MS = 3e3;
967
984
  function emptyLedger() {
@@ -2890,7 +2907,8 @@ function contract(name) {
2890
2907
  if (c?.decorationId === void 0) throw new Error(`job contract '${name}' is missing`);
2891
2908
  return { path: c.path, decorationId: c.decorationId };
2892
2909
  }
2893
- var PAGE_SIZE2 = 25;
2910
+ var PAGE_SIZE2 = 100;
2911
+ var BATCH_SIZE = 100;
2894
2912
  function queryTuple(q2) {
2895
2913
  const locationUnion = q2.geoId !== void 0 ? { geoId: q2.geoId } : q2.location !== void 0 ? { seoLocation: { location: q2.location } } : void 0;
2896
2914
  const selected = Object.entries(q2.filters).filter(([, v]) => v.length > 0);
@@ -2914,6 +2932,10 @@ function jobDetailUrl(id) {
2914
2932
  const c = contract("jobPosting");
2915
2933
  return `${c.path}/${id}?decorationId=${c.decorationId}`;
2916
2934
  }
2935
+ function jobBatchUrl(ids) {
2936
+ const c = contract("jobPosting");
2937
+ return `${c.path}?ids=List(${ids.join(",")})&decorationId=${c.decorationId}`;
2938
+ }
2917
2939
  function jobIdOf(input) {
2918
2940
  const s = input.trim();
2919
2941
  if (/^\d+$/.test(s)) return s;
@@ -3022,7 +3044,7 @@ function parseJobCards(json) {
3022
3044
  return { ok: true, jobs, total, start, ...resolvedGeo(included), unknownTypes };
3023
3045
  }
3024
3046
  var WORKPLACE_URN = { "1": "on-site", "2": "remote", "3": "hybrid" };
3025
- function applyOf(method) {
3047
+ function applyOf(method, id) {
3026
3048
  const m = method;
3027
3049
  if (m === null || m === void 0) return void 0;
3028
3050
  const type = String(m.$type ?? "");
@@ -3031,19 +3053,15 @@ function applyOf(method) {
3031
3053
  return defined2({ type: "external", url });
3032
3054
  }
3033
3055
  if (type.endsWith("OnsiteApply")) {
3034
- const url = typeof m.easyApplyUrl === "string" ? m.easyApplyUrl : void 0;
3035
- return defined2({ type: "easy-apply", url });
3056
+ const url = typeof m.easyApplyUrl === "string" ? m.easyApplyUrl : viewUrl(id);
3057
+ return { type: "easy-apply", url };
3036
3058
  }
3037
3059
  return void 0;
3038
3060
  }
3039
- function parseJobDetail(json) {
3040
- const root = json;
3041
- const d = root?.data;
3042
- const rawId = d?.jobPostingId ?? d?.entityUrn;
3061
+ function postingToJob(d, included) {
3062
+ const rawId = d.jobPostingId ?? d.entityUrn;
3043
3063
  const id = rawId === void 0 ? void 0 : jobIdOf(String(rawId));
3044
- if (d === void 0 || id === void 0 || typeof d.title !== "string")
3045
- return drift2("the job response carried no posting");
3046
- const included = root?.included ?? [];
3064
+ if (id === void 0 || typeof d.title !== "string") return void 0;
3047
3065
  const find = (urn) => typeof urn === "string" ? included.find((n) => n.entityUrn === urn) : void 0;
3048
3066
  const details = d.companyDetails ?? {};
3049
3067
  const company = find(details.company ?? details["*companyResolutionResult"]);
@@ -3053,31 +3071,53 @@ function parseJobDetail(json) {
3053
3071
  const applying = find(d["*applyingInfo"]);
3054
3072
  const saving = find(d["*savingInfo"]);
3055
3073
  const list2 = (v) => Array.isArray(v) && v.length > 0 ? v.filter((x) => typeof x === "string") : void 0;
3056
- return {
3057
- ok: true,
3058
- job: defined2({
3059
- id,
3060
- title: d.title,
3061
- company: companyName,
3062
- companyUrl: typeof company?.url === "string" ? company.url : void 0,
3063
- location: typeof d.formattedLocation === "string" ? d.formattedLocation : void 0,
3064
- workplace: workplace ?? (d.workRemoteAllowed === true ? "remote" : void 0),
3065
- employmentType: typeof d.formattedEmploymentStatus === "string" && d.formattedEmploymentStatus !== "" ? d.formattedEmploymentStatus : void 0,
3066
- experienceLevel: typeof d.formattedExperienceLevel === "string" && d.formattedExperienceLevel !== "" ? d.formattedExperienceLevel : void 0,
3067
- industries: list2(d.formattedIndustries),
3068
- functions: list2(d.formattedJobFunctions),
3069
- listed: day(d.listedAt),
3070
- expires: day(d.expireAt),
3071
- state: typeof d.jobState === "string" ? d.jobState : void 0,
3072
- applicants: typeof d.applies === "number" && d.applies > 0 ? d.applies : void 0,
3073
- apply: applyOf(d.applyMethod),
3074
- ats: typeof d.applicantTrackingSystem === "string" ? d.applicantTrackingSystem : void 0,
3075
- applied: typeof applying?.applied === "boolean" ? applying.applied : void 0,
3076
- saved: typeof saving?.saved === "boolean" ? saving.saved : void 0,
3077
- description: text2(d.description),
3078
- url: viewUrl(id)
3079
- })
3080
- };
3074
+ const nonEmpty = (v) => typeof v === "string" && v !== "" ? v : void 0;
3075
+ return defined2({
3076
+ id,
3077
+ title: d.title,
3078
+ company: companyName,
3079
+ companyUrl: typeof company?.url === "string" ? company.url : void 0,
3080
+ location: nonEmpty(d.formattedLocation),
3081
+ workplace: workplace ?? (d.workRemoteAllowed === true ? "remote" : void 0),
3082
+ employmentType: nonEmpty(d.formattedEmploymentStatus),
3083
+ experienceLevel: nonEmpty(d.formattedExperienceLevel),
3084
+ industries: list2(d.formattedIndustries),
3085
+ functions: list2(d.formattedJobFunctions),
3086
+ listed: day(d.listedAt),
3087
+ expires: day(d.expireAt),
3088
+ state: nonEmpty(d.jobState),
3089
+ applicants: typeof d.applies === "number" && d.applies > 0 ? d.applies : void 0,
3090
+ apply: applyOf(d.applyMethod, id),
3091
+ ats: nonEmpty(d.applicantTrackingSystem),
3092
+ applied: typeof applying?.applied === "boolean" ? applying.applied : void 0,
3093
+ saved: typeof saving?.saved === "boolean" ? saving.saved : void 0,
3094
+ description: text2(d.description),
3095
+ url: viewUrl(id)
3096
+ });
3097
+ }
3098
+ function parseJobDetail(json) {
3099
+ const root = json;
3100
+ const job = root?.data === void 0 ? void 0 : postingToJob(root.data, root.included ?? []);
3101
+ return job === void 0 ? drift2("the job response carried no posting") : { ok: true, job };
3102
+ }
3103
+ function parseJobBatch(json) {
3104
+ const root = json;
3105
+ const results = root?.data?.results;
3106
+ if (results === null || typeof results !== "object")
3107
+ return drift2("the batch job response carried no results");
3108
+ const included = root?.included ?? [];
3109
+ const byUrn = new Map(included.map((n) => [n.entityUrn, n]));
3110
+ const jobs = [];
3111
+ for (const ref of Object.values(results)) {
3112
+ const node = typeof ref === "string" ? byUrn.get(ref) : ref;
3113
+ const job = node === void 0 ? void 0 : postingToJob(node, included);
3114
+ if (job !== void 0) jobs.push(job);
3115
+ }
3116
+ const claimed = Object.keys(results).length;
3117
+ if (claimed > 0 && jobs.length === 0)
3118
+ return drift2(`the batch listed ${claimed} jobs but none could be read`);
3119
+ const errors = root?.data?.errors ?? {};
3120
+ return { ok: true, jobs, missing: Object.keys(errors).map((k) => k.replace(/^\*/, "")) };
3081
3121
  }
3082
3122
  function parseJobFilters(json) {
3083
3123
  const elements = json?.data?.elements;
@@ -3258,7 +3298,7 @@ function queryOf(req, filters) {
3258
3298
  async function collect(get, query, first, want) {
3259
3299
  const out = { jobs: [], exhausted: false, unknownTypes: {} };
3260
3300
  for (let start = first; out.jobs.length < want && start < MAX_RESULTS; start += PAGE_SIZE2) {
3261
- const res = await get(jobSearchUrl({ ...query, start, count: PAGE_SIZE2 }), "search");
3301
+ const res = await get(jobSearchUrl({ ...query, start, count: PAGE_SIZE2 }), "jobSearch");
3262
3302
  const page = res.ok ? parseJobCards(res.json) : res;
3263
3303
  if (!page.ok) {
3264
3304
  if (out.jobs.length === 0) return page;
@@ -3278,15 +3318,29 @@ async function collect(get, query, first, want) {
3278
3318
  }
3279
3319
  return out;
3280
3320
  }
3281
- async function addDetails(get, jobs, withDescription) {
3282
- for (const job of jobs) {
3283
- const res = await get(jobDetailUrl(job.id), "profile");
3284
- const detail = res.ok ? parseJobDetail(res.json) : res;
3285
- if (!detail.ok) return detail.code;
3286
- const { description, id: _id, url: _url, ...rest } = detail.job;
3287
- Object.assign(job, rest, withDescription ? { description } : {});
3321
+ async function readJobs(get, ids) {
3322
+ const jobs = /* @__PURE__ */ new Map();
3323
+ for (let at = 0; at < ids.length; at += BATCH_SIZE) {
3324
+ const res = await get(jobBatchUrl(ids.slice(at, at + BATCH_SIZE)), "jobRead");
3325
+ const batch = res.ok ? parseJobBatch(res.json) : res;
3326
+ if (!batch.ok) return { jobs, stoppedBy: batch.code };
3327
+ for (const job of batch.jobs) if (ids.includes(job.id)) jobs.set(job.id, job);
3288
3328
  }
3289
- return void 0;
3329
+ return { jobs };
3330
+ }
3331
+ async function addDetails(get, rows, withDescription) {
3332
+ const read = await readJobs(
3333
+ get,
3334
+ rows.map((r) => r.id)
3335
+ );
3336
+ for (const row of rows) {
3337
+ const detail = read.jobs.get(row.id);
3338
+ if (detail === void 0) continue;
3339
+ const { description, id: _id, url: _url, ...rest } = detail;
3340
+ Object.assign(row, rest, withDescription ? { description } : {});
3341
+ }
3342
+ const unread = rows.filter((r) => !read.jobs.has(r.id)).length;
3343
+ return read.stoppedBy === void 0 ? { unread } : { unread, stoppedBy: read.stoppedBy };
3290
3344
  }
3291
3345
  async function runJobs(req, opts = {}, fetch2) {
3292
3346
  const hasQuery = [req.keywords, req.location, req.geoId].some((v) => v !== void 0 && v !== "");
@@ -3308,9 +3362,11 @@ async function runJobs(req, opts = {}, fetch2) {
3308
3362
  const query = queryOf(req, selection2.filters);
3309
3363
  const found = await collect(get, query, first, want);
3310
3364
  if ("ok" in found) return err("jobs", found.code, found.message, found.hint);
3365
+ let unread = 0;
3311
3366
  if (req.details === true) {
3312
- const stopped = await addDetails(get, found.jobs, req.description === true);
3313
- found.stoppedBy = found.stoppedBy ?? stopped;
3367
+ const added = await addDetails(get, found.jobs, req.description === true);
3368
+ found.stoppedBy = found.stoppedBy ?? added.stoppedBy;
3369
+ unread = added.unread;
3314
3370
  }
3315
3371
  const { jobs, stoppedBy, exhausted, unknownTypes } = found;
3316
3372
  return ok("jobs", {
@@ -3318,29 +3374,59 @@ async function runJobs(req, opts = {}, fetch2) {
3318
3374
  ...found.resolved !== void 0 ? { location: found.resolved } : {},
3319
3375
  items: shape(jobs, opts, compactJob),
3320
3376
  meta: {
3321
- state: stoppedBy === void 0 && exhausted ? "complete" : "partial",
3377
+ state: stoppedBy === void 0 && exhausted && unread === 0 ? "complete" : "partial",
3322
3378
  operation: "jobs",
3323
3379
  claimedCount: found.total,
3324
3380
  returnedCount: jobs.length,
3325
3381
  start: first,
3326
3382
  next: exhausted ? void 0 : first + jobs.length,
3327
3383
  ...stoppedBy !== void 0 ? { stoppedBy } : {},
3384
+ ...unread > 0 ? { unread } : {},
3328
3385
  ...Object.keys(unknownTypes).length > 0 ? { unknownTypes } : {}
3329
3386
  }
3330
3387
  });
3331
3388
  }
3332
3389
  async function runJob(input, opts = {}, fetch2) {
3333
- const id = input === void 0 ? void 0 : jobIdOf(input);
3334
- if (id === void 0) {
3335
- return err("job", "INVALID_INPUT", "a job id or job URL is required", "lnrelay job <id|url>");
3390
+ const given = input === void 0 ? [] : Array.isArray(input) ? input : [input];
3391
+ const ids = [...new Set(given.map(jobIdOf))];
3392
+ if (ids.length === 0 || ids.some((id) => id === void 0)) {
3393
+ const bad = given.filter((g) => jobIdOf(g) === void 0);
3394
+ return err(
3395
+ "job",
3396
+ "INVALID_INPUT",
3397
+ bad.length > 0 ? `not a job id or URL: ${bad.join(", ")}` : "a job id or job URL is required",
3398
+ "lnrelay job <id|url> [<id|url> ...]"
3399
+ );
3336
3400
  }
3337
3401
  const get = fetch2 ?? liveFetch("job", opts.quiet === true);
3338
3402
  if (typeof get !== "function") return get;
3339
- const res = await get(jobDetailUrl(id), "profile");
3340
- if (!res.ok) return err("job", res.code, res.message, res.hint);
3341
- const detail = parseJobDetail(res.json);
3342
- if (!detail.ok) return err("job", detail.code, detail.message, detail.hint);
3343
- return ok("job", detail.job);
3403
+ const wanted = ids;
3404
+ if (wanted.length === 1) {
3405
+ const res = await get(jobDetailUrl(wanted[0]), "jobRead");
3406
+ if (!res.ok) return err("job", res.code, res.message, res.hint);
3407
+ const detail = parseJobDetail(res.json);
3408
+ if (!detail.ok) return err("job", detail.code, detail.message, detail.hint);
3409
+ return ok("job", detail.job);
3410
+ }
3411
+ const read = await readJobs(get, wanted);
3412
+ if (read.jobs.size === 0 && read.stoppedBy !== void 0)
3413
+ return err("job", read.stoppedBy, "no job could be read");
3414
+ const unread = wanted.filter((id) => !read.jobs.has(id));
3415
+ const jobs = wanted.flatMap((id) => read.jobs.get(id) ?? []);
3416
+ return ok("job", {
3417
+ items: shape(
3418
+ jobs,
3419
+ opts,
3420
+ (j) => compactJob({ ...j, easyApply: j.apply?.type === "easy-apply", promoted: false })
3421
+ ),
3422
+ meta: {
3423
+ state: unread.length === 0 ? "complete" : "partial",
3424
+ operation: "job",
3425
+ returnedCount: jobs.length,
3426
+ ...unread.length > 0 ? { unread } : {},
3427
+ ...read.stoppedBy !== void 0 ? { stoppedBy: read.stoppedBy } : {}
3428
+ }
3429
+ });
3344
3430
  }
3345
3431
  async function runJobFilters(req, opts = {}, fetch2) {
3346
3432
  const selection2 = jobFilterSelection(req.filter ?? {});
@@ -3348,7 +3434,7 @@ async function runJobFilters(req, opts = {}, fetch2) {
3348
3434
  const get = fetch2 ?? liveFetch("job-filters", opts.quiet === true);
3349
3435
  if (typeof get !== "function") return get;
3350
3436
  const url = jobFiltersUrl({ ...queryOf(req, selection2.filters), start: 0, count: PAGE_SIZE2 });
3351
- const res = await get(url, "search");
3437
+ const res = await get(url, "jobSearch");
3352
3438
  const parsed = res.ok ? parseJobFilters(res.json) : res;
3353
3439
  if (!parsed.ok) return err("job-filters", parsed.code, parsed.message, parsed.hint);
3354
3440
  return ok("job-filters", {
@@ -3597,7 +3683,15 @@ function runBudget(now, resetCooldown, confirmed) {
3597
3683
  `moved to ${corrupt.quarantinedTo}. A corrupt ledger is NOT treated as an empty one \u2014 that would silently restore a full budget. Inspect it, then re-run.`
3598
3684
  );
3599
3685
  }
3600
- const classes = ["profile", "search", "page", "write", "global"];
3686
+ const classes = [
3687
+ "profile",
3688
+ "search",
3689
+ "page",
3690
+ "jobSearch",
3691
+ "jobRead",
3692
+ "write",
3693
+ "global"
3694
+ ];
3601
3695
  const { value: cooldown } = loadCooldown();
3602
3696
  return ok("budget", {
3603
3697
  classes: classes.map((c) => ({
@@ -4024,7 +4118,7 @@ var COMMANDS = [
4024
4118
  },
4025
4119
  {
4026
4120
  name: "jobs",
4027
- cost: "1 search call per 25 jobs; --details adds 1 profile call per job",
4121
+ cost: "1 job-search call per 100 jobs; --details adds 1 job read per 100 jobs",
4028
4122
  summary: "Job search with every filter the jobs page has. Cards carry id, title, company, place, workplace, listed date, Easy Apply. Read finalists with `job`.",
4029
4123
  usage: 'lnrelay jobs "<keywords>" [--location "<place>" | --geo <id>] [--distance <miles>]\n [--posted 24h|week|month|<N>h|<N>d] [--experience internship,entry,associate,\n mid-senior,director,executive] [--type full-time,part-time,contract,temporary,\n volunteer,internship,other] [--workplace on-site,remote,hybrid] [--easy-apply]\n [--under-10] [--verified] [--in-network] [--fair-chance] [--company <ids>]\n [--industry <ids>] [--function <codes>] [--title <ids>] [--place <geo-ids>]\n [--filter key=v1,v2 ...] [--sort recent|relevant] [--limit N] [--start N]\n [--details [--description]] [--compact | --fields a,b,c]\n Ids for --company/--industry/--title/--place/--filter come from `job-filters`.\n The resolved place is in data.location: check it, "Vienna" may resolve narrow.',
4030
4124
  audience: ["cli", "mcp"],
@@ -4033,16 +4127,16 @@ var COMMANDS = [
4033
4127
  },
4034
4128
  {
4035
4129
  name: "job",
4036
- cost: "1 profile call",
4037
- summary: "One job: description, apply link (external or Easy Apply), applicants, state.",
4038
- usage: "lnrelay job <id|job-url>",
4130
+ cost: "1 job read per 100 jobs",
4131
+ summary: "Jobs in full: description, apply link (external or Easy Apply), applicants, state. Several ids are read together, 100 a call.",
4132
+ usage: "lnrelay job <id|job-url> [<id|job-url> ...] [--compact | --fields a,b,c]",
4039
4133
  audience: ["cli", "mcp"],
4040
4134
  risk: "read",
4041
4135
  implemented: true
4042
4136
  },
4043
4137
  {
4044
4138
  name: "job-filters",
4045
- cost: "1 search call",
4139
+ cost: "1 job-search call",
4046
4140
  summary: "Every job filter for a query, with option ids and live counts: what `jobs --filter` and --company/--industry/--title/--place take.",
4047
4141
  usage: 'lnrelay job-filters "<keywords>" [--location "<place>" | --geo <id>] [jobs filters]',
4048
4142
  audience: ["cli", "mcp"],
@@ -6706,7 +6800,7 @@ async function dispatch(argv, now) {
6706
6800
  case "jobs":
6707
6801
  return runJobs(jobsRequest(args), output(args));
6708
6802
  case "job":
6709
- return runJob(args.positionals[0], output(args));
6803
+ return runJob(args.positionals, output(args));
6710
6804
  case "job-filters":
6711
6805
  return runJobFilters(jobsRequest(args), output(args));
6712
6806
  case "search":