Add country-agnostic linkedin-search skill (#20)

A general-purpose, field-agnostic job-search skill built on LinkedIn's public jobs-guest endpoints. Works for any market out of the box — location is an explicit required flag (no country default). Zero runtime dependencies (bun only); search + detail commands. Includes a personal-use / Terms-of-Service note (automated access is against LinkedIn's ToS — keep volume low, non-commercial). (Pairs with the .gitignore fix in #21, which lets skills under .agents/ be tracked.) Co-authored-by: Akhil Tripathi <kodabear@Akhils-MacBook-Pro.local> Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
2026-09-17 00:26:26 +00:00 · 2026-06-29 20:26:19 +02:00
parent 46a9fdc66c
commit d8f38fe766
10 changed files with 724 additions and 1 deletions
@@ -0,0 +1,98 @@
+---
+name: linkedin-search
+version: 1.0.0
+description: >
+  Use this skill whenever the user wants to search for jobs in any location or
+  market, find job listings, or look up a specific job posting — in any country,
+  city, or remotely. Invoke for open positions, vacancies, and hiring across any
+  sector or role (software, data, design, marketing, finance, legal, operations,
+  etc.). The location is always supplied explicitly by the user. Trigger phrases:
+  find a job, job search, search for jobs, job openings, vacancies, hiring,
+  positions open, remote jobs, "are there any X jobs in <place>", look up this
+  job posting.
+context: fork
+allowed-tools: Bash(bun run skills/linkedin-search/cli/src/cli.ts *)
+---
+
+# LinkedIn Search Skill
+
+Search live job listings from LinkedIn's public job board for **any country/region**
+(and remote). No authentication, no API key, and **zero runtime dependencies** — it runs
+with just `bun`. The location is always passed explicitly, so the same skill works for a
+forker in any market out of the box.
+
+> This is a country-agnostic worked example of the repo's job-portal-skill pattern.
+> LinkedIn's `jobs-guest` endpoints are global and the HTML parsing is country-independent;
+> only the `--location` you pass changes per market.
+
+## ⚠️ Personal use only
+
+This uses LinkedIn's public job pages; automated access is against LinkedIn's Terms of
+Service, so **keep volume low and don't use it commercially or for bulk data collection.**
+Run it on your own responsibility.
+
+## When to use this skill
+
+- Search for job openings in a given location (any country/city) or remotely
+- Filter by recency (posted today / last 7 / 14 / 30 days) or workplace type (remote/hybrid/onsite)
+- Get the full description of a specific job listing
+
+## Commands
+
+### Search job listings
+
+```bash
+bun run skills/linkedin-search/cli/src/cli.ts search --location "<place>" [flags]
+```
+
+Key flags:
+- `--location <text>` / `-l <text>` — **required.** A LinkedIn place string, e.g. `"Mumbai, Maharashtra, India"`, `"Berlin, Germany"`, `"London, United Kingdom"`, or `"Remote"`.
+- `--query <text>` / `-q <text>` — keyword search (title, skill, role). Recommended.
+- `--jobage <days>` — posted within N days: `1`, `7`, `14`, `30`. Omit for all postings.
+- `--remote <mode>` — `remote`, `hybrid`, or `onsite` (workplace-type filter).
+- `--page <n>` — page number (1-indexed, 10 results per page).
+- `--limit <n>` / `-n <n>` — cap total results emitted (client-side).
+- `--format json|table|plain` — default `json`.
+
+### Fetch full job detail
+
+```bash
+bun run skills/linkedin-search/cli/src/cli.ts detail <id|url> [--format json|plain]
+```
+
+`id` is the job ID from `search` results (e.g. `4426311357`). You may also pass a full
+LinkedIn `jobs/view/...` URL or a `urn:li:jobPosting:...` URN. Returns the full description,
+seniority, employment type, job function, industries, and apply link.
+
+## Usage examples
+
+```bash
+# Data engineer roles in Bengaluru, last 30 days
+bun run skills/linkedin-search/cli/src/cli.ts search -q "data engineer" -l "Bengaluru, Karnataka, India" --jobage 30 --format table
+
+# Product manager roles in Berlin, remote
+bun run skills/linkedin-search/cli/src/cli.ts search -q "product manager" -l "Berlin, Germany" --remote remote --format table
+
+# Any role, fully remote
+bun run skills/linkedin-search/cli/src/cli.ts search -q "paralegal" -l "Remote" --format table
+
+# Full details for a specific job
+bun run skills/linkedin-search/cli/src/cli.ts detail 4426311357 --format plain
+```
+
+## Output formats
+
+| Format | Best for |
+|--------|----------|
+| `json` | Default — programmatic use, passing IDs to `detail` |
+| `table` | Quick human-readable scanning |
+| `plain` | Reading a single job's full detail (`detail` command) |
+
+All errors are written to **stderr** as `{ "error": "...", "code": "..." }` and the process exits with code `1`.
+
+## Notes
+
+- Data is from LinkedIn's public `jobs-guest` endpoints — no credentials required.
+- Page size is fixed at 10 results per page.
+- LinkedIn may rate-limit; the CLI retries 429/5xx with exponential backoff. Keep volume low (see ToS note above).
+- Job IDs are numeric (e.g. `4426311357`) — pass them as-is to `detail`.
@@ -0,0 +1,61 @@
+# linkedin-cli
+
+CLI for searching jobs on LinkedIn's public job listings, for **any country/region**
+(and remote), across any sector.
+
+**Data source**: LinkedIn `jobs-guest` endpoints (`seeMoreJobPostings/search` and `jobPosting/<id>`).
+**Authentication**: None required.
+**Dependencies**: None (plain `bun` + `fetch`). `bun install` is optional and only pulls dev type defs.
+
+> **Personal use only.** This uses LinkedIn's public job pages; automated access is against
+> LinkedIn's Terms of Service. Keep volume low, don't use it commercially or for bulk data
+> collection, and run it on your own responsibility.
+
+## Installation
+
+```bash
+cd .agents/skills/linkedin-search/cli
+bun install   # optional — only installs TypeScript dev types
+```
+
+The CLI runs without any install because it has zero runtime dependencies.
+
+## Commands
+
+| Command | Description |
+|---------|-------------|
+| `search` | Search for job listings (`--location` required) |
+| `detail` | Fetch full detail for a single job listing |
+
+`search` accepts `--format json|table|plain` (default `json`); `detail` accepts `--format json|plain`.
+All errors are written to **stderr** as `{ "error": "...", "code": "..." }` with exit code `1`.
+
+## Quick examples
+
+```bash
+# Software roles in Hyderabad, last 7 days
+bun run src/cli.ts search -q "backend engineer" -l "Hyderabad, Telangana, India" --jobage 7 --format table
+
+# Design roles in London
+bun run src/cli.ts search -q "product designer" -l "London, United Kingdom" --format table
+
+# Fully remote
+bun run src/cli.ts search -q "technical writer" -l "Remote" --remote remote --format table
+
+# Full detail for one job
+bun run src/cli.ts detail 4426311357 --format plain
+```
+
+See `../SKILL.md` for the full flag reference and the Terms-of-Service note.
+
+## Search flags
+
+| Flag | Alias | Description |
+|------|-------|-------------|
+| `--location` | `-l` | **Required.** Place string, e.g. `"Mumbai, Maharashtra, India"`, `"Berlin, Germany"`, `"Remote"`. |
+| `--query` | `-q` | Keywords (title / skill / role). Recommended. |
+| `--jobage` | | Posted within N days: `1`, `7`, `14`, `30`. |
+| `--remote` | | `remote` \| `hybrid` \| `onsite`. |
+| `--page` | | 1-indexed page (10 results/page). |
+| `--limit` | `-n` | Cap results emitted. |
+| `--format` | | `json` \| `table` \| `plain`. |
@@ -0,0 +1,19 @@
+{
+  "name": "linkedin-cli",
+  "version": "1.0.0",
+  "description": "CLI for searching jobs on LinkedIn's public job listings, for any country/region (and remote) — no authentication, zero runtime dependencies. Personal use only (LinkedIn ToS).",
+  "type": "module",
+  "main": "src/cli.ts",
+  "bin": {
+    "linkedin-search": "src/cli.ts"
+  },
+  "scripts": {
+    "start": "bun run src/cli.ts",
+    "typecheck": "tsc --noEmit"
+  },
+  "dependencies": {},
+  "devDependencies": {
+    "typescript": "^5.4.0",
+    "@types/bun": "latest"
+  }
+}
@@ -0,0 +1,116 @@
+#!/usr/bin/env bun
+// Self-contained CLI for searching jobs on LinkedIn's public jobs-guest endpoints,
+// for any country/region (plus remote). No external CLI framework, so it runs
+// anywhere `bun` is available with zero install beyond the repo clone.
+//
+// Personal use only. This reads LinkedIn's public job pages; automated access is
+// against LinkedIn's Terms of Service, so keep volume low and do not use it
+// commercially or for bulk data collection. Run it on your own responsibility.
+
+import { runSearch, type SearchOpts } from "./commands/search.js"
+import { runDetail, type DetailOpts } from "./commands/detail.js"
+
+interface Flags {
+  _: string[]
+  [k: string]: string | boolean | string[]
+}
+
+function parseFlags(argv: string[]): Flags {
+  const flags: Flags = { _: [] }
+  const alias: Record<string, string> = { q: "query", l: "location", n: "limit" }
+  for (let i = 0; i < argv.length; i++) {
+    const a = argv[i]
+    if (a.startsWith("--") || a.startsWith("-")) {
+      const key = alias[a.replace(/^-+/, "")] ?? a.replace(/^-+/, "")
+      const next = argv[i + 1]
+      if (next === undefined || next.startsWith("-")) {
+        flags[key] = true
+      } else {
+        flags[key] = next
+        i++
+      }
+    } else {
+      ;(flags._ as string[]).push(a)
+    }
+  }
+  return flags
+}
+
+const HELP = `linkedin-cli — search jobs on LinkedIn (any country/region, plus remote)
+
+USAGE
+  bun run src/cli.ts search --location "<place>" [flags]
+  bun run src/cli.ts detail <id|url> [--format json|plain]
+
+SEARCH FLAGS
+  --location, -l <text>   Location to search. REQUIRED. e.g. "Mumbai, Maharashtra, India",
+                          "Berlin, Germany", "London, United Kingdom", or "Remote".
+  --query, -q <text>      Keywords (job title, skill, or role). Recommended.
+  --jobage <days>         Posted within N days: 1, 7, 14, 30. Default: all.
+  --remote <mode>         remote | hybrid | onsite. Filter by workplace type.
+  --page <n>              1-indexed page (10 results/page). Default 1.
+  --limit, -n <n>         Cap results emitted (client-side).
+  --format <fmt>          json (default) | table | plain.
+
+EXAMPLES
+  bun run src/cli.ts search -q "data engineer" -l "Bengaluru, Karnataka, India" --jobage 30 --format table
+  bun run src/cli.ts search -q "product manager" -l "Berlin, Germany" --remote remote --format table
+  bun run src/cli.ts search -q "paralegal" -l "Remote" --format table
+  bun run src/cli.ts detail 4300011451 --format plain
+
+Personal use only — uses LinkedIn's public pages; keep volume low (LinkedIn ToS).
+`
+
+async function main(): Promise<number> {
+  const argv = process.argv.slice(2)
+  const flags = parseFlags(argv)
+  const cmd = (flags._ as string[])[0]
+
+  if (!cmd || flags.help || flags.h) {
+    process.stdout.write(HELP)
+    return cmd ? 0 : 1
+  }
+
+  if (cmd === "search") {
+    const location = typeof flags.location === "string" ? flags.location : undefined
+    if (!location) {
+      process.stderr.write(
+        JSON.stringify({
+          error: 'the --location/-l flag is required (e.g. -l "Mumbai, Maharashtra, India", -l "Berlin, Germany", or -l "Remote")',
+          code: "NO_LOCATION",
+        }) + "\n",
+      )
+      return 1
+    }
+    const fmt = (flags.format as string) || "json"
+    const opts: SearchOpts = {
+      query: typeof flags.query === "string" ? flags.query : undefined,
+      location,
+      jobage: flags.jobage ? parseInt(flags.jobage as string, 10) : 9999,
+      remote: typeof flags.remote === "string" ? flags.remote : undefined,
+      page: flags.page ? Math.max(1, parseInt(flags.page as string, 10)) : 1,
+      limit: flags.limit ? parseInt(flags.limit as string, 10) : undefined,
+      format: (["json", "table", "plain"].includes(fmt) ? fmt : "json") as SearchOpts["format"],
+    }
+    return runSearch(opts)
+  }
+
+  if (cmd === "detail") {
+    const id = (flags._ as string[])[1]
+    if (!id) {
+      process.stderr.write(JSON.stringify({ error: "detail requires an <id|url>", code: "NO_ID" }) + "\n")
+      return 1
+    }
+    const fmt = (flags.format as string) || "json"
+    const opts: DetailOpts = {
+      id,
+      format: (fmt === "plain" ? "plain" : "json") as DetailOpts["format"],
+    }
+    return runDetail(opts)
+  }
+
+  process.stderr.write(JSON.stringify({ error: `Unknown command "${cmd}"`, code: "BAD_CMD" }) + "\n")
+  return 1
+}
+
+main().then((code) => process.exit(code))
@@ -0,0 +1,57 @@
+import { DETAIL_URL, htmlFetch, parseJobDetail, writeError } from "../helpers.js"
+
+export interface DetailOpts {
+  id: string
+  format: "json" | "plain"
+}
+
+/** Accept a raw job ID, a job-view URL, or a job URN. */
+function normalizeId(input: string): string | null {
+  const urn = input.match(/urn:li:jobPosting:(\d+)/)
+  if (urn) return urn[1]
+  const url = input.match(/-(\d{6,})(?:\?|$)/) || input.match(/\/(\d{6,})(?:\?|$)/)
+  if (url) return url[1]
+  const bare = input.match(/^\d{6,}$/)
+  if (bare) return input
+  return null
+}
+
+export async function runDetail(opts: DetailOpts): Promise<number> {
+  const id = normalizeId(opts.id)
+  if (!id) {
+    writeError(`Could not parse a job ID from "${opts.id}"`, "BAD_ID")
+    return 1
+  }
+  try {
+    const html = await htmlFetch(`${DETAIL_URL}/${id}`)
+    if (!html) {
+      writeError("Job not found", "NOT_FOUND")
+      return 1
+    }
+    const job = parseJobDetail(html, id)
+
+    if (opts.format === "plain") {
+      const lines = [
+        job.title,
+        `${job.company || "—"} · ${job.location || "—"}`,
+        "",
+        job.seniority ? `Seniority: ${job.seniority}` : "",
+        job.employmentType ? `Employment: ${job.employmentType}` : "",
+        job.jobFunction ? `Function: ${job.jobFunction}` : "",
+        job.industries ? `Industries: ${job.industries}` : "",
+        "",
+        job.description || "(no description)",
+        "",
+        `URL: ${job.url}`,
+        job.applyUrl ? `Apply: ${job.applyUrl}` : "",
+      ].filter((l) => l !== "")
+      process.stdout.write(lines.join("\n") + "\n")
+    } else {
+      process.stdout.write(JSON.stringify(job, null, 2) + "\n")
+    }
+    return 0
+  } catch (e) {
+    writeError(e instanceof Error ? e.message : String(e), "DETAIL_FAILED")
+    return 1
+  }
+}
@@ -0,0 +1,85 @@
+import {
+  SEARCH_URL,
+  htmlFetch,
+  parseJobCards,
+  jobageToTPR,
+  workTypeFlag,
+  writeError,
+  type JobCard,
+} from "../helpers.js"
+
+export interface SearchOpts {
+  query?: string
+  location: string
+  jobage: number
+  remote?: string // "remote" | "hybrid" | "onsite"
+  page: number
+  limit?: number
+  format: "json" | "table" | "plain"
+}
+
+function buildUrl(opts: SearchOpts): string {
+  const params = new URLSearchParams()
+  if (opts.query) params.set("keywords", opts.query)
+  if (opts.location) params.set("location", opts.location)
+  const tpr = jobageToTPR(opts.jobage)
+  if (tpr) params.set("f_TPR", tpr)
+  const wt = workTypeFlag(opts.remote)
+  if (wt) params.set("f_WT", wt)
+  params.set("start", String((opts.page - 1) * 10))
+  return `${SEARCH_URL}?${params.toString()}`
+}
+
+function renderTable(cards: JobCard[]): string {
+  if (cards.length === 0) return "No results."
+  const rows = cards.map((c) => {
+    const title = (c.title || "").slice(0, 42).padEnd(42)
+    const company = (c.company || "—").slice(0, 26).padEnd(26)
+    const loc = (c.location || "—").slice(0, 24).padEnd(24)
+    const date = c.date || "—"
+    return `${c.id.padEnd(11)} ${title} ${company} ${loc} ${date}`
+  })
+  const header =
+    "ID".padEnd(11) +
+    " " +
+    "TITLE".padEnd(42) +
+    " " +
+    "COMPANY".padEnd(26) +
+    " " +
+    "LOCATION".padEnd(24) +
+    " DATE"
+  return [header, "-".repeat(header.length), ...rows].join("\n")
+}
+
+export async function runSearch(opts: SearchOpts): Promise<number> {
+  try {
+    const html = await htmlFetch(buildUrl(opts))
+    let cards = parseJobCards(html)
+    if (opts.limit && opts.limit > 0) cards = cards.slice(0, opts.limit)
+
+    if (opts.format === "table") {
+      process.stdout.write(renderTable(cards) + "\n")
+    } else if (opts.format === "plain") {
+      process.stdout.write(
+        cards
+          .map(
+            (c) =>
+              `${c.title}\n  ${c.company || "—"} · ${c.location || "—"} · ${c.date || "—"}\n  id: ${c.id}\n  ${c.url}`,
+          )
+          .join("\n\n") + "\n",
+      )
+    } else {
+      process.stdout.write(
+        JSON.stringify(
+          { meta: { count: cards.length, page: opts.page }, results: cards },
+          null,
+          2,
+        ) + "\n",
+      )
+    }
+    return 0
+  } catch (e) {
+    writeError(e instanceof Error ? e.message : String(e), "SEARCH_FAILED")
+    return 1
+  }
+}
@@ -0,0 +1,229 @@
+// Data source: LinkedIn public "jobs-guest" endpoints. No authentication required.
+// Search returns an HTML list of job cards; detail returns a single job's HTML.
+// We parse both with regex (the markup is shallow and stable; a full DOM parser
+// is unnecessary and node-html-parser has known nesting bugs on LinkedIn cards).
+
+export const SEARCH_URL =
+  "https://www.linkedin.com/jobs-guest/jobs/api/seeMoreJobPostings/search"
+export const DETAIL_URL =
+  "https://www.linkedin.com/jobs-guest/jobs/api/jobPosting"
+
+export function writeError(error: string, code: string): void {
+  process.stderr.write(JSON.stringify({ error, code }) + "\n")
+}
+
+const UA =
+  "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 " +
+  "(KHTML, like Gecko) Chrome/120.0.0.0 Safari/537.36"
+
+/** Fetch HTML with exponential backoff on 429/5xx. Returns "" on a 404. */
+export async function htmlFetch(url: string): Promise<string> {
+  const maxRetries = 6
+  let delay = 500
+  for (let attempt = 0; attempt <= maxRetries; attempt++) {
+    const response = await fetch(url, {
+      headers: {
+        "User-Agent": UA,
+        Accept: "text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8",
+        "Accept-Language": "en-US,en;q=0.9",
+        "X-Requested-With": "XMLHttpRequest",
+      },
+      redirect: "follow",
+    })
+    if (response.status === 429 || response.status >= 500) {
+      if (attempt === maxRetries) {
+        throw new Error(`Request failed: ${response.status} ${response.statusText}`)
+      }
+      const jitter = Math.floor(Math.random() * 500)
+      await new Promise((r) => setTimeout(r, delay + jitter))
+      delay = Math.min(delay * 2, 8000)
+      continue
+    }
+    if (response.status === 404) return ""
+    if (!response.ok) {
+      throw new Error(`Request failed: ${response.status} ${response.statusText}`)
+    }
+    return response.text()
+  }
+  throw new Error("Request failed after max retries")
+}
+
+export interface JobCard {
+  id: string
+  title: string
+  company: string | null
+  companyUrl: string | null
+  location: string | null
+  date: string | null
+  url: string
+}
+
+export interface JobDetail extends JobCard {
+  description: string | null
+  seniority: string | null
+  employmentType: string | null
+  jobFunction: string | null
+  industries: string | null
+  applyUrl: string | null
+}
+
+function decodeHtmlEntities(text: string): string {
+  return text
+    .replace(/&amp;/g, "&")
+    .replace(/&lt;/g, "<")
+    .replace(/&gt;/g, ">")
+    .replace(/&quot;/g, '"')
+    .replace(/&#39;/g, "'")
+    .replace(/&apos;/g, "'")
+    .replace(/&#(\d+);/g, (_, code) => String.fromCharCode(parseInt(code, 10)))
+    .replace(/&nbsp;/g, " ")
+}
+
+function stripTags(html: string): string {
+  return html.replace(/<[^>]+>/g, " ").replace(/\s+/g, " ").trim()
+}
+
+function clean(html: string): string {
+  return decodeHtmlEntities(stripTags(html))
+}
+
+/** Parse the job ID out of a LinkedIn job-view URL or URN. */
+function idFromUrl(url: string): string | null {
+  const m = url.match(/-(\d{6,})(?:\?|$)/) || url.match(/(\d{6,})/)
+  return m ? m[1] : null
+}
+
+/**
+ * Parse the search response: a flat list of <li> job cards. We split on the
+ * job-posting URN and parse each chunk independently so one malformed card
+ * cannot break the rest.
+ */
+export function parseJobCards(html: string): JobCard[] {
+  const results: JobCard[] = []
+  const chunks = html.split(/data-entity-urn="urn:li:jobPosting:/).slice(1)
+
+  for (const chunk of chunks) {
+    const idMatch = chunk.match(/^(\d+)/)
+    if (!idMatch) continue
+    const id = idMatch[1]
+
+    // Full link + title (title lives in the sr-only span or the <h3> title).
+    const linkMatch = chunk.match(/class="base-card__full-link[^"]*"[^>]*href="([^"]+)"/i)
+    const url = linkMatch ? decodeHtmlEntities(linkMatch[1]).split("?")[0] : ""
+
+    let title: string | null = null
+    const h3 = chunk.match(/class="base-search-card__title"[^>]*>([\s\S]*?)<\/h3>/i)
+    if (h3) title = clean(h3[1])
+    if (!title) {
+      const sr = chunk.match(/class="sr-only"[^>]*>([\s\S]*?)<\/span>/i)
+      if (sr) title = clean(sr[1])
+    }
+    if (!title) continue
+
+    // Company (subtitle <h4> with optional inner <a>).
+    let company: string | null = null
+    let companyUrl: string | null = null
+    const sub = chunk.match(/class="base-search-card__subtitle"[^>]*>([\s\S]*?)<\/h4>/i)
+    if (sub) {
+      const a = sub[1].match(/href="([^"]+)"/i)
+      if (a) companyUrl = decodeHtmlEntities(a[1]).split("?")[0]
+      company = clean(sub[1]) || null
+    }
+
+    // Location + date.
+    const loc = chunk.match(/class="job-search-card__location"[^>]*>([\s\S]*?)<\/span>/i)
+    const location = loc ? clean(loc[1]) || null : null
+    const dt = chunk.match(/class="job-search-card__listdate[^"]*"[^>]*datetime="([^"]+)"/i)
+    const date = dt ? dt[1] : null
+
+    results.push({
+      id,
+      title,
+      company,
+      companyUrl,
+      location,
+      date,
+      url: url || `https://www.linkedin.com/jobs/view/${id}`,
+    })
+  }
+
+  return results
+}
+
+/** Parse the single-job detail page. */
+export function parseJobDetail(html: string, id: string): JobDetail {
+  const title = html.match(
+    /class="(?:top-card-layout__title|topcard__title)[^"]*"[^>]*>([\s\S]*?)<\/h[12]>/i,
+  )?.[1]
+  const orgMatch = html.match(
+    /class="topcard__org-name-link[^"]*"[^>]*href="([^"]+)"[^>]*>([\s\S]*?)<\/a>/i,
+  )
+  const company = orgMatch ? clean(orgMatch[2]) || null : null
+  const companyUrl = orgMatch ? decodeHtmlEntities(orgMatch[1]).split("?")[0] : null
+
+  const locMatch = html.match(
+    /class="topcard__flavor topcard__flavor--bullet"[^>]*>([\s\S]*?)<\/span>/i,
+  )
+  const location = locMatch ? clean(locMatch[1]) || null : null
+
+  // Rich description block. Keep paragraph/line breaks as newlines.
+  let description: string | null = null
+  const desc = html.match(
+    /class="(?:show-more-less-html__markup|description__text[^"]*)"[^>]*>([\s\S]*?)<\/div>/i,
+  )
+  if (desc) {
+    const withBreaks = desc[1]
+      .replace(/<\s*br\s*\/?>/gi, "\n")
+      .replace(/<\/(p|li|ul|ol|div|h\d)>/gi, "\n")
+    description = decodeHtmlEntities(stripTags(withBreaks)).replace(/\n{3,}/g, "\n\n").trim() || null
+  }
+
+  // Job-criteria items: subheader label -> text value.
+  const criteria: Record<string, string> = {}
+  const itemRe =
+    /class="description__job-criteria-subheader"[^>]*>([\s\S]*?)<\/h3>[\s\S]*?class="description__job-criteria-text[^"]*"[^>]*>([\s\S]*?)<\/span>/gi
+  let cm: RegExpExecArray | null
+  while ((cm = itemRe.exec(html)) !== null) {
+    criteria[clean(cm[1]).toLowerCase()] = clean(cm[2])
+  }
+
+  const applyMatch = html.match(/class="topcard__link[^"]*"[^>]*href="([^"]+)"/i)
+  const applyUrl = applyMatch ? decodeHtmlEntities(applyMatch[1]).split("?")[0] : null
+
+  return {
+    id,
+    title: title ? clean(title) : "(untitled)",
+    company,
+    companyUrl,
+    location,
+    date: null,
+    url: `https://www.linkedin.com/jobs/view/${id}`,
+    description,
+    seniority: criteria["seniority level"] ?? null,
+    employmentType: criteria["employment type"] ?? null,
+    jobFunction: criteria["job function"] ?? null,
+    industries: criteria["industries"] ?? null,
+    applyUrl,
+  }
+}
+
+/** Convert a job-age in days to LinkedIn's f_TPR seconds value. */
+export function jobageToTPR(days: number): string | null {
+  if (!days || days <= 0 || days >= 9999) return null
+  return `r${days * 86400}`
+}
+
+/** Workplace-type flag: on-site=1, remote=2, hybrid=3. */
+export function workTypeFlag(mode: string | undefined): string | null {
+  switch ((mode || "").toLowerCase()) {
+    case "remote":
+      return "2"
+    case "hybrid":
+      return "3"
+    case "onsite":
+    case "on-site":
+      return "1"
+    default:
+      return null
+  }
+}
@@ -0,0 +1,14 @@
+{
+  "compilerOptions": {
+    "target": "ESNext",
+    "module": "ESNext",
+    "moduleResolution": "bundler",
+    "types": ["bun"],
+    "strict": true,
+    "noEmit": true,
+    "skipLibCheck": true,
+    "allowImportingTsExtensions": true,
+    "verbatimModuleSyntax": true
+  },
+  "include": ["src/**/*.ts"]
+}
@@ -0,0 +1,42 @@
+# LinkedIn Jobs URL Reference
+
+Public, unauthenticated `jobs-guest` endpoints used by this skill. Global — the same
+endpoints serve every market; only the `location` value changes.
+
+> Personal use only — automated access is against LinkedIn's Terms of Service; keep volume low.
+
+## Search
+
+```
+GET https://www.linkedin.com/jobs-guest/jobs/api/seeMoreJobPostings/search
+```
+
+Query params:
+
+| Param | Meaning | Example |
+|-------|---------|---------|
+| `keywords` | Free-text query | `data engineer` |
+| `location` | Place string | `Mumbai, Maharashtra, India` · `Berlin, Germany` · `Remote` |
+| `f_TPR` | Posted-within window (seconds) | `r604800` (7d), `r2592000` (30d) |
+| `f_WT` | Workplace type | `1` on-site · `2` remote · `3` hybrid |
+| `start` | Pagination offset (10/page) | `0`, `10`, `20`, … |
+
+Returns an HTML list of job cards (one `<li>` per posting). The CLI parses each card by
+its `data-entity-urn="urn:li:jobPosting:<id>"` and extracts title, company, location, date, URL.
+
+## Detail
+
+```
+GET https://www.linkedin.com/jobs-guest/jobs/api/jobPosting/<jobId>
+```
+
+Returns a single job's HTML: title (`top-card-layout__title`), company
+(`topcard__org-name-link`), location (`topcard__flavor--bullet`), the rich description
+(`show-more-less-html__markup` / `description__text`), and job-criteria items
+(seniority, employment type, job function, industries).
+
+## Notes
+
+- No authentication required.
+- Respect rate limits — the CLI backs off on 429/5xx.
+- Country-agnostic: pass any `--location` (city, region, country, or "Remote").
@@ -198,7 +198,9 @@ The CV uses [moderncv](https://ctan.org/pkg/moderncv) (banking style). The cover

 ### Job search tools

-The four CLI tools in `.agents/skills/` are specific to the **Danish job market** (Jobbank, Jobdanmark, Jobindex, Jobnet). They demonstrate the pattern for building job portal integrations. If you're in a different country, you can build equivalent tools for your local job portals using the same structure.
+The four Danish CLI tools in `.agents/skills/` (Jobbank, Jobdanmark, Jobindex, Jobnet) demonstrate the pattern for building a job-portal integration for a specific market. If you're in a different country, you can build equivalent tools for your local job portals using the same structure.
+
+For a **country-agnostic** starting point, the repo also includes **`linkedin-search`** — a job-search skill built on LinkedIn's public, unauthenticated `jobs-guest` endpoints. It is field-agnostic, has **zero runtime dependencies** (runs with just `bun`), and takes the search location as an explicit flag, so it works for any market out of the box (`-l "Berlin, Germany"`, `-l "Mumbai, Maharashtra, India"`, `-l "Remote"`, …). It is intended for **personal use only** — automated access is against LinkedIn's Terms of Service, so keep volume low. See `.agents/skills/linkedin-search/SKILL.md`.

 ### Salary benchmarking