# jobs

> Reference for the jobs tool: search the boards, list stored postings, save to the pipeline, read the morning digest.

Source: https://careers.geenius.io/docs/reference/jobs/

The jobs tool searches the public boards, lists what is stored (best match first) and holds the saved pipeline.

Call it as `geenius_careers(tool: "jobs", action: "<action>", args: {...})` through Geenius OS.

## Actions

| Action | What it does |
|---|---|
| `search` | Fetch new freelance and contract postings from the public boards, filter, dedupe, store and score them against the CV. Returns the run's reports and the best new matches. |
| `list` | Stored postings, newest first, each with its match score, its parsed office policy (mode, days per week or month) and its materials: CV version, cover letter and e-mail drafts. |
| `save` | Save a posting to the pipeline (stage saved). Nothing is drafted or sent. |
| `get` | One posting read back with its materials: the CV version with a link, the cover letter, and every e-mail draft with what will be attached, its approval request and the pipeline stage. |
| `saved` | The application pipeline: saved postings with their stage (drafted, applied, interview and so on) and their materials. |
| `digest` | The morning digest: strong matches since a time, as Markdown. |
| `since_digest` | What is new since the person's last daily digest: that digest and every strong match stored after it, best first, each with its job id. Also says whether the daily digest is on. |

## Arguments we have checked

### jobs · search

Fetch new freelance and contract postings from the public boards, filter, dedupe, store and score them against the CV. Returns the run's reports and the best new matches.

- `query`: Words to search for.
- `sources`: A list of board names.
- `freelance_only`: Only freelance and contract postings.
- `eu_only`: Only postings open to people in the EU.
- `top`: How many best matches to return.

### jobs · list

Stored postings, newest first, each with its match score, its parsed office policy (mode, days per week or month) and its materials: CV version, cover letter and e-mail drafts.

- `query`: Filters title, company and location.
- `min_score`: Keep only strong matches.
- `limit`: Default 30, maximum 100.
- `near`: Places where any office load is fine; each row then says whether it fits.
- `max_office_days_per_week`: The most office days you accept.

### jobs · save

Save a posting to the pipeline (stage saved). Nothing is drafted or sent.

- `job_id`: A stored posting.
- `posting`: A posting you found yourself: source, title, url and optionally company, location, kind, rateText, description. It is stored first and deduplicated on its address.

### jobs · get

One posting read back with its materials: the CV version with a link, the cover letter, and every e-mail draft with what will be attached, its approval request and the pipeline stage.

- `job_id` (required): The posting.

### jobs · digest

The morning digest: strong matches since a time, as Markdown.

- `since`: Epoch milliseconds. Default 24 hours ago.
- `min_score`: Default 75.
- `limit`: How many to list.

### jobs · since_digest

What is new since the person's last daily digest: that digest and every strong match stored after it, best first, each with its job id. Also says whether the daily digest is on.

- `min_score`: Minimum score.
- `limit`: How many to list.

> [!TIP]
> Ask for the exact arguments of any action with `action: "help", of: "search"`. The server answers from its own definitions, so that answer is always current.