API

Every tisane.api.* operation is listed below with parameter shapes and return types from the plugin Zod schemas.

New to Corsair? See API access, authentication, and error handling.

Text

extractEntities

text.extractEntities Extract named entities and topics from text Risk: read

await corsair.tisane.api.text.extractEntities({});

Input

Name	Type	Required	Description
`content`	`string`	Yes	Text to extract named entities and topics from
`language`	`string`	Yes	—

Output

Name	Type	Required	Description
`entities`	`object[]`	Yes	—
`topics`	`object[]`	Yes	—

entities full type

{
  name: string,
  type?: any,
  subtypes?: any[],
  wikidata?: string,
  mentions?: any[]
}[]

topics full type

{
  topic: string,
  coverage?: number
}[]

moderate

text.moderate Content moderation for abuse, hate speech, and harassment Risk: read

await corsair.tisane.api.text.moderate({});

Input

Name	Type	Required	Description
`content`	`string`	Yes	Text to check for abusive language, hate speech, or harassment
`language`	`string`	Yes	—

Output

Name	Type	Required	Description
`abuse`	`object[]`	Yes	—
`flagged`	`boolean`	Yes	—

abuse full type

{
  type: string,
  severity?: string,
  text?: string,
  offset?: number,
  length?: number,
  sentence_index?: number,
  explanation?: string
}[]

parse

text.parse Comprehensive natural language parsing and text analysis Risk: read

await corsair.tisane.api.text.parse({});

Input

Name	Type	Required	Description
`content`	`string`	Yes	The text content to analyze
`language`	`string`	Yes	ISO 639-1 language code
`settings`	`object`	No	—

settings full type

{
  snippets?: boolean,
  sentiment?: boolean,
  entities?: boolean,
  topics?: boolean,
  abuse?: boolean,
  format?: string,
  explain?: boolean,
  topic_stats?: boolean
}

Output

Name	Type	Required	Description
`text`	`string`	Yes	—
`language`	`string`	No	—
`topics`	`object[]`	No	—
`abuse`	`object[]`	No	—
`entities_summary`	`object[]`	No	—
`sentiment_expressions`	`object[]`	No	—

topics full type

(
  string | {
    topic: string,
    coverage?: number
  }
)[]

abuse full type

{
  type: string,
  severity?: string,
  text?: string,
  offset?: number,
  length?: number,
  sentence_index?: number,
  explanation?: string
}[]

entities_summary full type

{
  name: string,
  type?: any,
  subtypes?: any[],
  wikidata?: string,
  mentions?: any[]
}[]

sentiment_expressions full type

{
  polarity: positive | negative | mixed | neutral,
  offset?: number,
  length?: number,
  sentence_index?: number,
  targets?: any[],
  reasons?: any[],
  text?: string,
  explanation?: string
}[]

sentiment

text.sentiment Aspect-based sentiment analysis and tone evaluation Risk: read

await corsair.tisane.api.text.sentiment({});

Input

Name	Type	Required	Description
`content`	`string`	Yes	Text to analyze for aspect-based sentiment
`language`	`string`	Yes	—

Output

Name	Type	Required	Description
`sentiment`	`object[]`	Yes	—
`text`	`string`	No	—

sentiment full type

{
  polarity: positive | negative | mixed | neutral,
  offset?: number,
  length?: number,
  sentence_index?: number,
  targets?: any[],
  reasons?: any[],
  text?: string,
  explanation?: string
}[]

Overview

Plugins

Text

extractEntities

moderate

parse

sentiment

​Text

​extractEntities

​moderate

​parse

​sentiment

Text

extractEntities

moderate

parse

sentiment