JobsPipe

JobsPipe is a developer API with two functions over one host and one key:

1. Job search — live, normalized job postings from 30+ ATS and job-board

sources (Workday, Greenhouse, Lever, Ashby, LinkedIn, and more), deduplicated into one JSON schema.

1. Tech search (stack scan) — detect the technologies a given domain runs

(frameworks, CDNs, analytics, payments, and more), with confidence scores.

Base URL: https://api.jobspipe.dev

When to use

• Find, list, or analyze open job postings (by title, skill/tech, location,

remote, seniority, company, recency).

• Use open roles as a hiring/buying signal for a company.
• Identify which technologies a company or domain uses.

Authentication

Every request needs an API key sent as a Bearer token. Keys start with jp_live_. Get one free at https://jobspipe.dev/signup (free tier: 5,000 requests/month, no card).

Authorization: Bearer jp_live_YOUR_KEY

Set it once:

export JOBSPIPE_API_KEY=jp_live_YOUR_KEY

Install the CLI

Installing this skill does not install the CLI — npx skills add only copies this file. Install the CLI once (global), so the jobspipe command is on PATH, then set your key:

npm i -g jobspipe-cli
export JOBSPIPE_API_KEY=jp_live_YOUR_KEY

Then jobspipe jobs ... and jobspipe stack ... work directly. Run jobspipe --help for every flag; verify your key with jobspipe jobs --limit 1. Can't install globally? Prefix any command with npx -y, e.g. npx -y jobspipe-cli jobs --limit 1. No Node at all? Use the curl examples below.

Function 1 — Job search

POST /v1/jobs/search. The body is a JSON object of filters; all fields are optional and combine with AND. Array filters ending in _or match any value. An empty body returns the most recent postings.

curl -s https://api.jobspipe.dev/v1/jobs/search \
  -H "Authorization: Bearer $JOBSPIPE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "job_title_or": ["software engineer", "backend engineer"],
    "job_country_code_or": ["US"],
    "remote": true,
    "posted_at_max_age_days": 7,
    "limit": 25,
    "include_total_results": true
  }'

CLI equivalent:

jobspipe jobs --title "software engineer" --title "backend engineer" \
  --country US --remote --max-age-days 7 --limit 25 --total

Pass a raw filter object for the full surface (overrides any flags, also read from stdin):

jobspipe jobs '{"description_or":["rust"],"posted_at_max_age_days":7,"limit":10}'

Common filters: job_title_or, job_title_not, description_or (match a skill or tech in the description), description_not, job_country_code_or, remote, posted_at_max_age_days, posted_at_gte and posted_at_lte (YYYY-MM-DD), company_name_or, company_name_partial_match_or, job_seniority_or, employment_type_or (full-time, part-time, contract, temporary, internship), source_or, limit, offset, include_total_results.

Response:

{
  "metadata": {
    "total_results": 412,
    "truncated_results": 25,
    "next_cursor": "eyJvZmZzZXQiOjI1fQ"
  },
  "data": [
    {
      "id": "jp_3958211043",
      "job_title": "Backend Engineer",
      "company": "Acme Inc",
      "company_domain": "acme.com",
      "location": "Remote · United States",
      "country_code": "US",
      "remote": true,
      "seniority": "Mid-Senior",
      "min_annual_salary_usd": 150000,
      "max_annual_salary_usd": 190000,
      "date_posted": "2026-06-18",
      "technology_slugs": ["go", "postgres"],
      "final_url": "https://example.com/careers/3958211043"
    }
  ]
}

limit is capped by your plan (free 25, builder 100, scale 500). Paginate with offset (or follow metadata.next_cursor) until fewer than limit rows return.

Function 2 — Tech search (stack scan)

POST /v1/stack/scan. Give a domain; get the technologies detected on it. mode is optional: auto (default, fast HTTP then headless render if thin), html, or render.

curl -s https://api.jobspipe.dev/v1/stack/scan \
  -H "Authorization: Bearer $JOBSPIPE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{ "domain": "stripe.com" }'

CLI equivalent:

jobspipe stack stripe.com
jobspipe stack vercel.com --mode render

Response:

{
  "domain": "stripe.com",
  "scanned_at": "2026-06-25T12:00:00.000Z",
  "http_status": 200,
  "render_path": "curl_cffi",
  "detected": [
    {
      "slug": "react",
      "name": "React",
      "categories": ["JavaScript frameworks"],
      "confidence": 95,
      "version": null,
      "signals": [{ "kind": "script", "match": "react.production.min.js" }],
      "website": "https://react.dev",
      "pricing": ["open source"],
      "saas": false,
      "oss": true
    }
  ]
}

Scans are cached for about 14 days; a repeat within that window is served from cache and still counts as one request.

Errors

Both functions return { "error": "..." } with these statuses:

• 400 — invalid domain (stack scan only).
• 401 — missing or invalid API key.
• 402 — monthly request quota exceeded.
• 429 — per-second rate limit exceeded; slow down and retry.
• 502 — scanner failed (stack scan); retry or try --mode render.
• 504 — upstream timed out; retry.

Notes

• Job records are deduplicated across sources; the same role appears once.
• Salary is parsed into structured ranges (*_annual_salary*) when present.
• Full schema and guides: https://jobspipe.dev/docs