> ## Documentation Index
> Fetch the complete documentation index at: https://docs.brussle.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Starters that read one document

> Ready-made judgments whose answer reads only the judged document, filled in with your own paths.

These starters answer from the judged document alone, so a column in your own database may be enough for them. You create each one as [starter judgments](/guides/starter-judgments#create-a-judgment-from-a-starter) shows, and whether an answer is good on your data is yours to measure against outcomes.

## Ticket triage: is it urgent?

For support queues that must show the tickets that cannot wait first. The judgment reads the ticket's subject and first message, and the customer's plan if you have one. When the conversation is an array, it also reads the latest three messages. It does not read the status, the assignee or the tags, so a change to them costs nothing. Sort or get alerts on `p` or on the `urgent` threshold.

`from_starter`: `ticket_triage.urgency`

| Placeholder | What it is | Example |
| - | - | - |
| `subject`, path | The ticket's subject line. | `state.subject` |
| `body`, path | The ticket's first message or description. | `state.body` |
| `messages`, path, optional | The conversation, as an array of messages. The judgment reads the latest three. | `state.messages` |
| `plan`, path, optional | The customer's plan or tier. | `attributes.plan` |

```json theme={"theme":{"light":"css-variables","dark":"css-variables"}}
{
  "name": "urgent",
  "type": "bool",
  "question": "Is this ticket urgent: does the customer need a reply within hours rather than days?",
  "criteria": "Urgent when a service is down or unusable, money is being lost or charged wrongly, data may be lost or exposed, or the customer faces a deadline today. A question, a feature request or a minor inconvenience is not urgent, however it is worded.",
  "context": {
    "fields": ["{subject}", "{body}", "{plan}"],
    "last_n": {"{messages}": 3},
    "max_tokens": 1500
  },
  "thresholds": {"urgent": 0.8}
}
```

## Ticket triage: which team should take it?

For routing tickets to a team. The judgment reads the same parts of a ticket as the urgency starter, so Brussle answers the two together. It picks billing, technical, account or sales, or none\_of\_the\_above when none fits. To use your own teams, change the options in the definition that the dry run returns. Use `value` as the answer. Until you post outcomes, the probabilities of a choice are too confident.

`from_starter`: `ticket_triage.routing`

| Placeholder | What it is | Example |
| - | - | - |
| `subject`, path | The ticket's subject line. | `state.subject` |
| `body`, path | The ticket's first message or description. | `state.body` |
| `messages`, path, optional | The conversation, as an array of messages. The judgment reads the latest three. | `state.messages` |
| `plan`, path, optional | The customer's plan or tier. | `attributes.plan` |

```json theme={"theme":{"light":"css-variables","dark":"css-variables"}}
{
  "name": "route",
  "type": "choice",
  "question": "Which team should handle this ticket?",
  "options": [
    {
      "value": "billing",
      "description": "Charges, invoices, refunds, payment methods and plan changes."
    },
    {
      "value": "technical",
      "description": "Bugs, errors, outages, performance and integrations."
    },
    {
      "value": "account",
      "description": "Signing in, access, users, permissions and account settings."
    },
    {
      "value": "sales",
      "description": "Pricing questions, upgrades, new purchases and contracts."
    }
  ],
  "context": {
    "fields": ["{subject}", "{body}", "{plan}"],
    "last_n": {"{messages}": 3},
    "max_tokens": 1500
  }
}
```

## Moderation: does this post break a rule?

For posts, comments and reviews that users write. The judgment reads the post's text and title. The criteria list common community rules. To use your own rules, change the criteria in the definition that the dry run returns. A change to likes, views or replies costs nothing. Hide a post above the `hide` threshold. Send a post between `review` and `hide` to a person.

To also show the post that a post replies to, add a relation. See the referenced document guide.

`from_starter`: `moderation.breaks_rule`

| Placeholder | What it is | Example |
| - | - | - |
| `text`, path | The post's text. | `state.text` |
| `title`, path, optional | The post's title, if posts have one. | `state.title` |

```json theme={"theme":{"light":"css-variables","dark":"css-variables"}}
{
  "name": "breaks_rule",
  "type": "bool",
  "question": "Does this post break any of the community's rules?",
  "criteria": "A post breaks a rule if it harasses or threatens a person, attacks people for who they are, is spam or a scam, shares someone's private information, or sexualises minors. Strong language, criticism, disagreement and off-topic posts do not break a rule on their own.",
  "context": {"fields": ["{title}", "{text}"], "max_tokens": 1000},
  "thresholds": {"hide": 0.9, "review": 0.5}
}
```

## Review queue: does a person need to look at this?

For a flow where your system handles most items and a person must look at some. Examples are refund requests, automated replies, applications and claims. The judgment reads the item's content, and its summary and category if you have them. Send the items above the `review` threshold to your queue. To review more items, make the threshold lower.

`from_starter`: `review_queue.needs_review`

| Placeholder | What it is | Example |
| - | - | - |
| `content`, path | The item itself: the request, reply or application. | `state.content` |
| `summary`, path, optional | A short summary or title. | `state.summary` |
| `category`, path, optional | The item's category or type. | `attributes.category` |

```json theme={"theme":{"light":"css-variables","dark":"css-variables"}}
{
  "name": "needs_review",
  "type": "bool",
  "question": "Should a person review this item before it is acted on?",
  "criteria": "Review when the item is ambiguous or contradicts itself, involves money, legal matters, safety or access to an account, the person is upset, or an automated answer could easily be wrong. Routine, clear requests do not need review.",
  "context": {
    "fields": ["{summary}", "{category}", "{content}"],
    "max_tokens": 1500
  },
  "thresholds": {"review": 0.5}
}
```


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.