38 new applicants and 2 offers accepted last week
38 new applicants, 17 interviews, 3 offers and 2 hires across 4 open jobs. Support Specialist got only 4 applicants.
Every Monday: new applicants, interviews, offers and hires per open Greenhouse job for last week, against the week before.
Reads
You get
The card on the right is a sample. Yours shows your own numbers and updates in place every run.
What lands on your board
38 new applicants, 17 interviews, 3 offers and 2 hires across 4 open jobs. Support Specialist got only 4 applicants.
Greenhouse has reports for all of this, and building the weekly one for a leadership meeting still means picking the report, the date range and the jobs, then copying numbers into a doc. Founders hiring their first ten people want the same four numbers per role every week, without logging in.
The agent reads last week's applications, interviews and offers through Greenhouse's Harvest API v3, groups them by job and compares with the week before. Roles where nobody new entered the funnel, or where interviews piled up without offers, stand out at a glance.
POST https://auth.greenhouse.io/token with your Harvest v3 client ID and secret (basic auth) and grant_type=client_credentials. Then list jobs with GET /v3/jobs and keep the open ones with their names.
GET /v3/applications?created_at[gte]=<last Monday, ISO 8601>&created_at[lt]=<this Monday>&prospect=false&per_page=500, following the Link header cursor (the cursor must be the only parameter on later pages). Group by job_id.
GET /v3/interviews?starts_at[gte]=<last Monday>&starts_at[lt]=<this Monday> and keep interviews that started before this Monday with status scheduled, awaiting_feedback, collect_feedback or complete. Map each application_id to its job.
GET /v3/offers?created_at[gte]=<last Monday>&created_at[lt]=<this Monday> for offers created, ignoring Deprecated versions; an offer with status Accepted and resolved_at last week counts as a hire.
One row per open job with applicants, interviews, offers and hires, totals with the weekly change, and a line on jobs with no new applicants or with many interviews and no offers.
With BRIEFKIN_TOKEN set, the agent sends the result as one card that updates in place every run. Without it, you get the report in the chat.
Pick your agent, where the data comes from and how to schedule it; the commands below follow.
One command. It puts the SKILL.md in ~/.claude/skills and in ~/.agents/skills, which Codex, Cursor, Gemini CLI, Copilot CLI and OpenCode read too.
shellcurl -fsSL https://briefkin.dev/skills/greenhouse-weekly-recruiting-report/install | shAsk Claude Code:
promptRun the greenhouse-weekly-recruiting-report skillOr paste the prompt instead of installing:
promptUsing the Greenhouse Harvest API v3 (read only, OAuth client credentials), report on last week, Monday to Sunday. List open jobs. Count new applications created last week per job, excluding prospects; interviews that took place last week per job (map application_id to job); offers created last week, ignoring Deprecated versions; and hires as offers Accepted last week. Do the same for the week before. Give totals with the change, one row per open job with applicants, interviews, offers and hires, and point out jobs with no new applicants or with three or more interviews and no offer.Mondays at 8:00, from cron on any machine where Claude Code is signed in:
crontab0 8 * * 1 claude -p "Run the greenhouse-weekly-recruiting-report skill"Add it with crontab -e. A laptop asleep at that time runs it when it wakes if you use launchd or Task Scheduler instead; GitHub Actions works for the API route.
Connect Briefkin once with curl -fsSL https://briefkin.dev/install | sh and every run lands on your board as a report card. How setup works
SKILL.md---
name: greenhouse-weekly-recruiting-report
description: "Every Monday: new applicants, interviews, offers and hires per open Greenhouse job for last week, against the week before. Use when the user asks for this, or when it runs on a schedule."
---
# Greenhouse weekly recruiting report
Every Monday: new applicants, interviews, offers and hires per open Greenhouse job for last week, against the week before.
Greenhouse has reports for all of this, and building the weekly one for a leadership meeting still means picking the report, the date range and the jobs, then copying numbers into a doc. Founders hiring their first ten people want the same four numbers per role every week, without logging in.
The agent reads last week's applications, interviews and offers through Greenhouse's Harvest API v3, groups them by job and compares with the week before. Roles where nobody new entered the funnel, or where interviews piled up without offers, stand out at a glance.
## Two ways to get the data
Use the API or MCP route when you have an MCP server, a CLI or an API token for the app. Otherwise, if you can drive a browser the user is signed in to (Claude in Chrome, a Playwright or Chrome DevTools MCP server), use the browser route. Say which one you used.
## Route A: API or MCP server
### What you need
- Harvest v3 API credentials (Configure > Dev Center > API Credential Management > Create New API Credentials, API type Harvest V3 (OAuth)) with read scopes for jobs, applications, interviews and offers, in `GREENHOUSE_CLIENT_ID` and `GREENHOUSE_CLIENT_SECRET`. List endpoints act as a Site Admin user.
- Harvest v1 and v2 were removed on August 31, 2026, so older Harvest API keys no longer work for this.
If something is missing, say what and stop; do not invent numbers.
### Steps
1. **Get a token and the open jobs** (Greenhouse): `POST https://auth.greenhouse.io/token` with your Harvest v3 client ID and secret (basic auth) and `grant_type=client_credentials`. Then list jobs with `GET /v3/jobs` and keep the open ones with their names.
2. **Count new applicants** (Greenhouse): `GET /v3/applications?created_at[gte]=<last Monday, ISO 8601>&created_at[lt]=<this Monday>&prospect=false&per_page=500`, following the `Link` header cursor (the cursor must be the only parameter on later pages). Group by `job_id`.
3. **Count interviews** (Greenhouse): `GET /v3/interviews?starts_at[gte]=<last Monday>&starts_at[lt]=<this Monday>` and keep interviews that started before this Monday with `status` `scheduled`, `awaiting_feedback`, `collect_feedback` or `complete`. Map each `application_id` to its job.
4. **Count offers and hires** (Greenhouse): `GET /v3/offers?created_at[gte]=<last Monday>&created_at[lt]=<this Monday>` for offers created, ignoring `Deprecated` versions; an offer with `status` `Accepted` and `resolved_at` last week counts as a hire.
5. **Report**: One row per open job with applicants, interviews, offers and hires, totals with the weekly change, and a line on jobs with no new applicants or with many interviews and no offers.
## Route B: the browser
### What you need
- Signed in to Greenhouse as a Site Admin or a Job Admin with the **Can view reports** permission on all open jobs; reports only include jobs the user can see.
If something is missing, say what and stop; do not invent numbers.
### Steps
Start at https://app.greenhouse.io/dashboard. The user is signed in already; never type passwords or codes, and ask if a sign-in page appears.
1. **Count new applicants per job**: Open **Reports** and the **Pipeline History per Job** report (search the report library if it is not on the first page). Set the date range to last Monday through Sunday, filter **Job status** to Open, and read the **Applied** (new applications) column per job. Run it again for the week before.
2. **Count interviews**: Open the **Interview Calendar** or **Scheduled Interviews** report for the same two weeks and count interviews per job, leaving out cancelled ones. If your library only has the per-candidate view, group by the job column after exporting.
3. **Count offers and hires**: Open the **Offer Activity** report (or **Offers** under Reports) for the week: offers created per job, and offers marked **Accepted** last week as hires, with the start date. Repeat for the week before.
4. **Report**: One row per open job with applicants, interviews, offers and hires, totals with the weekly change, and a line on jobs with no new applicants or with three or more interviews and no offer. Use **Export** on each report when a table is longer than one page.
Note: Report names in the library differ slightly between Greenhouse plans and over time; the three reports to look for are the pipeline history per job, scheduled interviews and offer activity. Prospects are excluded from pipeline reports by default, which matches the API route.
## The report
- **Totals**: new applicants, interviews, offers and hires last week, with the change.
- **Per open job**: the same four numbers.
- **Stalled roles**: jobs with no new applicants, or interviews without offers.
- **Accepted offers**: role and start date.
## Post it to Briefkin
When `BRIEFKIN_TOKEN` is set, or `~/.config/briefkin/env` exists (the installer from https://briefkin.dev/docs writes it), also send the result to the user's Briefkin board as one card. Keep the `item_key` the same on every run so the card updates in place, use a new `event_id` per run, and put today's real numbers in place of the example ones. If the request fails, say so once and carry on.
```bash
[ -n "${BRIEFKIN_TOKEN:-}" ] || . ~/.config/briefkin/env
curl -sS -X POST "${BRIEFKIN_URL:-https://briefkin.dev}/v1/events" \
-H "authorization: Bearer $BRIEFKIN_TOKEN" \
-H "content-type: application/json" \
-d @- <<'JSON'
{
"v": 1,
"event_id": "greenhouse-weekly-recruiting-report/<date of this run, e.g. 2026-10-16>",
"item_key": "greenhouse-weekly-recruiting-report",
"project": "Acme Hiring",
"source": "Greenhouse weekly recruiting report skill",
"kind": "update",
"title": "38 new applicants and 2 offers accepted last week",
"summary": "**38** new applicants, **17** interviews, **3** offers and **2** hires across 4 open jobs. Support Specialist got only 4 applicants.",
"widget": [
{
"type": "Grid",
"props": {
"columns": 4
},
"children": [
{
"type": "Stat",
"props": {
"label": "Applicants",
"value": 38,
"delta": 15
}
},
{
"type": "Stat",
"props": {
"label": "Interviews",
"value": 17,
"delta": 13
}
},
{
"type": "Stat",
"props": {
"label": "Offers",
"value": 3
}
},
{
"type": "Stat",
"props": {
"label": "Hires",
"value": 2,
"tone": "ok"
}
}
]
},
{
"type": "Table",
"props": {
"columns": [
"Job",
"Applicants",
"Interviews",
"Offers",
"Hires"
],
"align": [
"left",
"right",
"right",
"right",
"right"
],
"rows": [
[
"Senior Backend Engineer",
14,
6,
1,
1
],
[
"Product Designer",
11,
4,
1,
0
],
[
"Account Executive",
9,
5,
1,
1
],
[
"Support Specialist",
4,
2,
0,
0
]
]
}
},
{
"type": "Callout",
"props": {
"text": "Support Specialist has had under 5 applicants for three weeks. The job post may need a refresh or a new board.",
"tone": "warn"
}
}
]
}
JSON
```
Widgets come from the Briefkin catalog (https://briefkin.dev/v1/catalog). Without a token, just give the user the report.
## Run it on a schedule
Suggested: Mondays at 8:00 (`0 8 * * 1`). From cron, with the agent signed in on that machine:
```bash
0 8 * * 1 claude -p "Run the greenhouse-weekly-recruiting-report skill"
# Codex: codex exec, Gemini CLI: gemini -p, Copilot CLI: copilot -p, OpenCode: opencode run
```
Or use the agent's own scheduler (Claude Code routines, Codex automations, Claude or ChatGPT scheduled tasks): https://briefkin.dev/skills/greenhouse-weekly-recruiting-report#schedule
Source: https://briefkin.dev/skills/greenhouse-weekly-recruiting-report
Harvest v3 API credentials (Configure > Dev Center > API Credential Management > Create New API Credentials, API type Harvest V3 (OAuth)) with read scopes for jobs, applications, interviews and offers, in GREENHOUSE_CLIENT_ID and GREENHOUSE_CLIENT_SECRET. List endpoints act as a Site Admin user.
Harvest v1 and v2 were removed on August 31, 2026, so older Harvest API keys no longer work for this.
GREENHOUSE_CLIENT_ID and GREENHOUSE_CLIENT_SECRET, exchanges them at POST https://auth.greenhouse.io/token with grant_type=client_credentials and basic auth, and sends the returned token as a Bearer token, fetching a new one when it expires.Give the credential read scopes only, for applications, jobs, stages, interviews, scorecards and offers. Candidate records hold personal data, so the cards the agent posts should name candidates and stages and leave everything else out.
Greenhouse removed Harvest v1 and v2 on August 31, 2026. The skill uses Harvest v3, which authenticates with OAuth client credentials: create new credentials in Greenhouse, choose Harvest V3 and grant read scopes. List endpoints act as a site admin user, so the credential needs one behind it.
Yes, every application on the job counts as an applicant whatever its source. If you want them split, ask for a breakdown by source_id and the agent will group applicants by source per job.
No. The skill runs with your own agent and gives you the report in the chat. With a Briefkin token set, each run also posts it to your board as a report card, so you can glance at it instead of asking again.
Claude Code, Codex, Cursor, Gemini CLI, GitHub Copilot CLI, OpenCode, Windsurf and Amp all read SKILL.md files. The install command puts it in ~/.claude/skills and ~/.agents/skills, where they look. ChatGPT, the Claude app and Grok take the prompt instead. Pick your agent under "Install and run it" and the commands change to match.
Yes. Switch the data source to "Your browser" and the steps change to the Greenhouse pages themselves: the agent opens app.greenhouse.io/dashboard in a browser you are signed in to and reads the numbers off the page. Claude Code, Codex, Cursor, Claude app and ChatGPT can drive a browser out of the box; Gemini CLI, Copilot CLI and OpenCode need a Playwright or Chrome DevTools MCP server first. It is slower than the API and needs you signed in, so it suits a report you run by hand or from a scheduler on your own computer.
Skills are free to use with any agent. Briefkin is where their results land: one board for every report, number and question your agents send, with a ping when something needs you.
Free during the open beta. No card.