# Headcount Labs

> Job postings search: find, count and read open jobs from employer career sites, with pay and links.

- name: com.headcountlabs/jobs
- version: 1.2.0
- connection class: R0 (autonomous)
- trust: 100/100
- quality: 84/100 (new)
- owner: not claimed
- descriptor: https://api.protogrid.dev/v1/servers/com.headcountlabs%2Fjobs
- tools: https://api.protogrid.dev/v1/servers/com.headcountlabs%2Fjobs/tools

**An agent can connect right now, no human step.** Remote endpoint, no authentication, reachable on the last probe.

## Connection (mcpServers)

```json
{
  "mcpServers": {
    "jobs": {
      "type": "http",
      "url": "https://api.headcountlabs.com/mcp/public"
    }
  }
}
```


## Trust

- hygiene: 100
- liveness: 100
- freshness: 100
- provenance: 100
- drivers: +repository +namespace-matches-repo +dns-namespace +website +reachable +uptime-100% +updated-2d-ago
- Derived from observable signals (official registry feed and our own credential-free probes); no code audit performed.

## Quality 84/100 (new)

### protocol: 75 (weight 25)

- warn `protocol.modern`: newest supported version is 2025-11-25; 2026-07-28 not supported, older versions are deprecated until 2027-07-28
- n/a `protocol.stateless`: only applies to 2026-07-28 servers
- pass `protocol.transport`: streamable HTTP
- n/a `protocol.list_ttl`: only applies to 2026-07-28 servers

### authorization: 83 (weight 15)

- pass `auth.prm`: Protected Resource Metadata (RFC 9728) published
- pass `auth.as_metadata`: authorization server metadata reachable
- warn `auth.cimd`: only Dynamic Client Registration, deprecated in 2026-07-28 in favour of CIMD
- n/a `auth.secret_in_url`: no templated URL

### tool hygiene: 93 (weight 30)

- pass `tools.descriptions`: every tool has a description
- warn `tools.description_length`: 2 tools have descriptions over 1,024 characters, which crowds the context
- pass `tools.schemas`: every tool has an object input schema
- pass `tools.annotations`: 9 of 9 tools declare readOnlyHint or destructiveHint
- pass `tools.directory_hints`: every tool declares readOnlyHint, destructiveHint, idempotentHint and openWorldHint
- pass `tools.token_cost`: about 7,167 tokens to load every tool
- pass `tools.api_dump`: tools are not a one-to-one API dump

### stability: n/a (weight 15)

- n/a `stability.changes`, `stability.rug_pull`: tool history recorded for 5 of the 7 days needed

### dependencies: n/a (weight 15)

- n/a `deps.known_vulns`, `deps.mcp_sdk_version`, `deps.resolvable`: no npm or PyPI package

- tools: 9, about 7,167 tokens to load them all
- tool set hash: 2b513c5eed9794bb96d1258dd0a5e33f
- tools last changed: 2026-10-20T06:13:28.017Z
- badge: [![protogrid quality](https://protogrid.dev/badge/com.headcountlabs/jobs.svg)](https://protogrid.dev/servers/com.headcountlabs/jobs)
- Checks run on what our credential-free, read-only probes observe; tools are never called and no code audit is performed.

### Tool changes

- 2026-10-20 `market_insights` removed
  - Answer job-market questions from a precomputed snapshot of current US jobs, rebuilt after each full pass over the job boards and checked by recomputing sample answers a second way. Ten ready-made questions (top employers, pay, remote share, top metros, jobs by state, top roles, new postings this week, pay transparency, posting age, employers hiring in the most states) plus a generic lookup of any metric (job count, pay 10th to 90th percentile, share showing pay, remote/hybrid/in-person shares, posting age, new postings, top employers) by one or two dimensions (title, place, company, seniority, employment type, remote), a breakdown that ranks one dimension, a negotiate question ("what salary should I ask for?" for a title, a place and an optional seniority) and a dimensions call that lists what can be asked. Cells with fewer than 30 jobs, and pay figures from fewer than 30 jobs with pay, are left out; every answer carries its sample size, as_of, a one-line method, the check result and search filters for search_jobs or count_jobs.
- 2026-10-20 `market_dimensions` added
  - List what market_lookup, market_breakdown and salary_range accept: the dimensions, how many values each has and the metrics; with dimension (and optionally q) the values of that dimension, such as the metro names. No account needed.
- 2026-10-20 `salary_range` added
  - What salary to ask for: for a title, a place (default US) and an optional seniority, the 25th, 50th, 75th and 90th percentile of posted pay, the sample size and a plain recommendation. Answers come from a daily snapshot of open US jobs and give their as_of date; for current job counts or the employers hiring right now, count_jobs is the live source. Cells with fewer than 30 jobs are left out. No account needed.
- 2026-10-20 `market_breakdown` added
  - Rank the values of one dimension (by: title, state, metro, city, company, seniority, employment_type or remote) by jobs, pay or another figure, over all jobs or inside one filter, such as which states pay Software Engineers the most (by=state, title=Software Engineer, sort=pay_median). Answers come from a daily snapshot of open US jobs and give their as_of date; for current job counts or the employers hiring right now, count_jobs is the live source. Cells with fewer than 30 jobs are left out. No account needed.
- 2026-10-20 `market_lookup` added
  - Job count, pay percentiles, share of jobs showing pay, remote/hybrid/in-person shares, posting age, new postings or top employers for up to two of title, place, company, seniority, employment_type and remote, such as the pay range for Registered Nurses in Texas. Leave metric out for every figure. Answers come from a daily snapshot of open US jobs and give their as_of date; for current job counts or the employers hiring right now, count_jobs is the live source. Cells with fewer than 30 jobs are left out. No account needed.
- 2026-10-20 `count_jobs` changed (description)
  - before: Count the current jobs matching the filters, without listing them. No account needed. To answer "which companies are hiring for X" or "how are these jobs split by ...", pass group_by: employer (company works too), work_arrangement (remote, hybrid or in_person), title, state (US), city (metro works too), country or employment_type. The answer then also has group_by, groups (the top 25 values by number of matching jobs, each with value and jobs), groups_from_jobs (how many matching jobs were grouped, at most 1,000, taken in the default list order), groups_sampled (true when more jobs matched than were grouped, so the counts are from a sample) and groups_unknown_jobs (grouped jobs with no known value). A job with several locations or postings can count under more than one value. group_by employer keeps its own keys: employers (each with employer_name, company_id and jobs), employers_from_jobs, employers_sampled and employers_unknown_jobs. When a filter is too broad to count exactly in time, count is an estimate: the answer then has count_is_estimate: true, count_margin (the 95% range either side of count) and count_note.
  - after: Count the current jobs matching the filters, without listing them. This is the live count: it reads the job database as it is now, so use it for "how many jobs" and "which employers are hiring" questions, and prefer its numbers over market_lookup and market_breakdown, which read a daily snapshot. No account needed. To answer "which companies are hiring for X" or "how are these jobs split by ...", pass group_by: employer (company works too), work_arrangement (remote, hybrid or in_person), title, state (US), city (metro works too), country or employment_type. The answer then also has group_by, groups (the top 25 values by number of matching jobs, each with value and jobs), groups_from_jobs (how many matching jobs were grouped, at most 1,000, taken in the default list order), groups_sampled (true when more jobs matched than were grouped, so the counts are from a sample) and groups_unknown_jobs (grouped jobs with no known value). A job with several locations or postings can count under more than one value. group_by employer keeps its own keys: employers (each with employer_name, company_id and jobs), employers_from_jobs, employers_sampled and employers_unknown_jobs. When a filter is too broad to count exactly in time, count is an estimate: the answer then has count_is_estimate: true, count_margin (the 95% range either side of count) and count_note.
- 2026-10-20 `list_fields` changed (description)
  - before: Find the search_jobs filters and the job fields that match some keywords, with each one's type and a short description. Pass words such as ["pay", "remote"]; no keywords lists everything. kind is filter (pass it in search_jobs filters) or field (a property of each returned job; postings[].x is a property of each linked posting). A field state is an object whose state says whether the source published the value (known), Atlas derived it, the source omitted it, or extraction failed. Free.
  - after: Find the search_jobs filters and the job fields that match some keywords, with each one's type and a short description. Pass words such as ["pay", "remote"]; no keywords lists everything. kind is filter (pass it in search_jobs filters) or field (a property of each returned job; postings[].x is a property of each linked posting). A field state is an object whose state says whether the source published the value (known), Atlas derived it, the source omitted it, or extraction failed.
- 2026-10-20 `market_insights` changed (description)
  - before: Answer job-market questions from a precomputed snapshot of current US jobs, rebuilt after each full pass over the job boards and checked by recomputing sample answers a second way. Ten ready-made questions (top employers, pay, remote share, top metros, jobs by state, top roles, new postings this week, pay transparency, posting age, employers hiring in the most states) plus a generic lookup of any metric (job count, pay 10th to 90th percentile, share showing pay, remote/hybrid/in-person shares, posting age, new postings, top employers) by one or two dimensions (title, place, company, seniority, employment type, remote), a breakdown that ranks one dimension, a negotiate question ("what salary should I ask for?" for a title, a place and an optional seniority) and a dimensions call that lists what can be asked. Cells with fewer than 30 jobs, and pay figures from fewer than 30 jobs with pay, are left out; every answer carries its sample size, as_of, a one-line method, the check result and search filters for search_jobs or count_jobs. Free.
  - after: Answer job-market questions from a precomputed snapshot of current US jobs, rebuilt after each full pass over the job boards and checked by recomputing sample answers a second way. Ten ready-made questions (top employers, pay, remote share, top metros, jobs by state, top roles, new postings this week, pay transparency, posting age, employers hiring in the most states) plus a generic lookup of any metric (job count, pay 10th to 90th percentile, share showing pay, remote/hybrid/in-person shares, posting age, new postings, top employers) by one or two dimensions (title, place, company, seniority, employment type, remote), a breakdown that ranks one dimension, a negotiate question ("what salary should I ask for?" for a title, a place and an optional seniority) and a dimensions call that lists what can be asked. Cells with fewer than 30 jobs, and pay figures from fewer than 30 jobs with pay, are left out; every answer carries its sample size, as_of, a one-line method, the check result and search filters for search_jobs or count_jobs.
- 2026-10-20 `search_jobs` changed (description)
  - before: Search current job postings by keyword, title, company, place, pay, date and more. No account needed. Call list_fields (free) to find the filter names. Country and employment_type accept comma-separated values or arrays and match any listed value. Pages return up to 50 jobs (limit, default 10), or up to 10 jobs when descriptions is "full"; a larger limit is refused. Pass next_cursor as cursor to see the next page. One search shows up to 250 jobs in total (the last page is shortened to fit), so narrow it with filters to find the right ones; count_jobs tells you how many jobs match without listing them. sort is relevance (best match) or updated (most recently updated). By default each job is a compact summary: rank, job_id, job_ad (the employer-written title, employer and a 200-character snippet of the description), company_id, locations, work_arrangement, pay (min, max, currency, period), posted_at, first_seen_at, posting_url and apply_url. For every field of each job pass view: "full" (descriptions cut to 400 characters, or descriptions: "full" for the whole text), name the top-level properties you need in fields (job_id is always included), or call get_job for one job. You choose how much description each job gets: descriptions: "snippet" (the default) gives a short preview and allows up to 50 jobs per page; descriptions: "full" gives the whole text but allows only 10 jobs per page. To read the full text of just a few jobs, search with snippets, then call get_job on the ones you want. Leave description out of fields to drop it entirely. For job titles, title matches the words as one phrase in that order, so "senior software engineer" misses "Software Engineer, Senior"; title_advanced matches words in any order with AND, OR and NOT, for example senior AND ("software engineer" OR "software developer"), and usually finds more.
  - after: Search current job postings by keyword, title, company, place, pay, date and more. No account needed. Call list_fields to find the filter names. Country and employment_type accept comma-separated values or arrays and match any listed value. Pages return up to 50 jobs (limit, default 10), or up to 10 jobs when descriptions is "full"; a larger limit is refused. Pass next_cursor as cursor to see the next page. One search shows up to 250 jobs in total (the last page is shortened to fit), so narrow it with filters to find the right ones; count_jobs tells you how many jobs match without listing them. sort is relevance (best match) or updated (most recently updated). By default each job is a compact summary: rank, job_id, job_ad (the employer-written title, employer and a 200-character snippet of the description), company_id, locations, work_arrangement, pay (min, max, currency, period), posted_at, first_seen_at, posting_url and apply_url. For every field of each job pass view: "full" (descriptions cut to 400 characters, or descriptions: "full" for the whole text), name the top-level properties you need in fields (job_id is always included), or call get_job for one job. You choose how much description each job gets: descriptions: "snippet" (the default) gives a short preview and allows up to 50 jobs per page; descriptions: "full" gives the whole text but allows only 10 jobs per page. To read the full text of just a few jobs, search with snippets, then call get_job on the ones you want. Leave description out of fields to drop it entirely. For job titles, title matches the words as one phrase in that order, so "senior software engineer" misses "Software Engineer, Senior"; title_advanced matches words in any order with AND, OR and NOT, for example senior AND ("software engineer" OR "software developer"), and usually finds more.
- 2026-10-20 `search_docs` changed (description)
  - before: Ask a question about Headcount Labs (filters, limits, MCP setup, errors) and get the best-matching sections of the headcountlabs.com docs, each with its URL. Free, and works without an account.
  - after: Ask a question about Headcount Labs (filters, limits, MCP setup, errors) and get the best-matching sections of the headcountlabs.com docs, each with its URL. Works without an account.

Full history: https://api.protogrid.dev/v1/servers/com.headcountlabs%2Fjobs/changes

## Remotes

- https://api.headcountlabs.com/mcp/public (streamable-http, auth none, reachable true, uptime30d 1)
- https://api.headcountlabs.com/mcp (streamable-http, auth oauth2, reachable true, uptime30d 1)

## Packages

_none_

## Tools (9)

- `count_jobs`: Count the current jobs matching the filters, without listing them. This is the live count: it reads the job database as it is now, so use it for "how many jobs" and "which employers are hiring" questions, and prefer its numbers over market_lookup and market_breakdown, which read a daily snapshot. No account needed. To answer "which companies are hiring for X" or "how are these jobs split by ...", pass group_by: employer (company works too), work_arrangement (remote, hybrid or in_person), title, state (US), city (metro works too), country or employment_type. The answer then also has group_by, groups (the top 25 values by number of matching jobs, each with value and jobs), groups_from_jobs (how many matching jobs were grouped, at most 1,000, taken in the default list order), groups_sampled (true when more jobs matched than were grouped, so the counts are from a sample) and groups_unknown_jobs (grouped jobs with no known value). A job with several locations or postings can count under more than one value. group_by employer keeps its own keys: employers (each with employer_name, company_id and jobs), employers_from_jobs, employers_sampled and employers_unknown_jobs. When a filter is too broad to count exactly in time, count is an estimate: the answer then has count_is_estimate: true, count_margin (the 95% range either side of count) and count_note.
- `get_job`: Get a job by its job_id. No account needed. By default the answer has the same compact shape search_jobs returns (rank 1, job_id, job_ad, company_id, locations, work_arrangement, pay, posted_at, first_seen_at, posting_url, apply_url) with the whole description in job_ad.description, currentness, stale and last_checked_at (is it still open, and when was it last checked), and every linked posting in postings (status, last_fetched_at, application_check). Pass view: "full" for every field as the REST API returns it. Use fields to select top-level job properties; job_id is always included.
- `list_fields`: Find the search_jobs filters and the job fields that match some keywords, with each one's type and a short description. Pass words such as ["pay", "remote"]; no keywords lists everything. kind is filter (pass it in search_jobs filters) or field (a property of each returned job; postings[].x is a property of each linked posting). A field state is an object whose state says whether the source published the value (known), Atlas derived it, the source omitted it, or extraction failed.
- `market_breakdown`: Rank the values of one dimension (by: title, state, metro, city, company, seniority, employment_type or remote) by jobs, pay or another figure, over all jobs or inside one filter, such as which states pay Software Engineers the most (by=state, title=Software Engineer, sort=pay_median). Answers come from a daily snapshot of open US jobs and give their as_of date; for current job counts or the employers hiring right now, count_jobs is the live source. Cells with fewer than 30 jobs are left out. No account needed.
- `market_dimensions`: List what market_lookup, market_breakdown and salary_range accept: the dimensions, how many values each has and the metrics; with dimension (and optionally q) the values of that dimension, such as the metro names. No account needed.
- `market_lookup`: Job count, pay percentiles, share of jobs showing pay, remote/hybrid/in-person shares, posting age, new postings or top employers for up to two of title, place, company, seniority, employment_type and remote, such as the pay range for Registered Nurses in Texas. Leave metric out for every figure. Answers come from a daily snapshot of open US jobs and give their as_of date; for current job counts or the employers hiring right now, count_jobs is the live source. Cells with fewer than 30 jobs are left out. No account needed.
- `salary_range`: What salary to ask for: for a title, a place (default US) and an optional seniority, the 25th, 50th, 75th and 90th percentile of posted pay, the sample size and a plain recommendation. Answers come from a daily snapshot of open US jobs and give their as_of date; for current job counts or the employers hiring right now, count_jobs is the live source. Cells with fewer than 30 jobs are left out. No account needed.
- `search_docs`: Ask a question about Headcount Labs (filters, limits, MCP setup, errors) and get the best-matching sections of the headcountlabs.com docs, each with its URL. Works without an account.
- `search_jobs`: Search current job postings by keyword, title, company, place, pay, date and more. No account needed. Call list_fields to find the filter names. Country and employment_type accept comma-separated values or arrays and match any listed value. Pages return up to 50 jobs (limit, default 10), or up to 10 jobs when descriptions is "full"; a larger limit is refused. Pass next_cursor as cursor to see the next page. One search shows up to 250 jobs in total (the last page is shortened to fit), so narrow it with filters to find the right ones; count_jobs tells you how many jobs match without listing them. sort is relevance (best match) or updated (most recently updated). By default each job is a compact summary: rank, job_id, job_ad (the employer-written title, employer and a 200-character snippet of the description), company_id, locations, work_arrangement, pay (min, max, currency, period), posted_at, first_seen_at, posting_url and apply_url. For every field of each job pass view: "full" (descriptions cut to 400 characters, or descriptions: "full" for the whole text), name the top-level properties you need in fields (job_id is always included), or call get_job for one job. You choose how much description each job gets: descriptions: "snippet" (the default) gives a short preview and allows up to 50 jobs per page; descriptions: "full" gives the whole text but allows only 10 jobs per page. To read the full text of just a few jobs, search with snippets, then call get_job on the ones you want. Leave description out of fields to drop it entirely. For job titles, title matches the words as one phrase in that order, so "senior software engineer" misses "Software Engineer, Senior"; title_advanced matches words in any order with AND, OR and NOT, for example senior AND ("software engineer" OR "software developer"), and usually finds more.