Models Reference¶

Complete schema definitions for all request and response models.

Request models¶

InputRequest¶

Used by POST /clearview.

Field	Type	Required	Description
`input`	`string`	Yes	A URL, file path, or raw text string to analyze

{
  "input": "https://example.com/article"
}

AudioRequest¶

Used by POST /audio.

Field	Type	Required	Description
`input`	`string`	Yes	A URL, file path, or raw text string to convert

{
  "input": "https://example.com/article"
}

Response models¶

ClearviewResponse¶

Returned by POST /clearview.

Field	Type	Description
`title`	`string`	Extracted document title
`content`	`string`	Full extracted text content
`source`	`string`	Original source URL or input path
`word_count`	`integer`	Number of words in the extracted content
`summary`	`string \\| null`	Auto-generated summary. `null` if summarization failed.
`scam_notes`	`string \\| null`	Human-readable explanation of the scam classification
`subjective_notes`	`string \\| null`	Human-readable explanation of the objectivity classification
`bias_notes`	`string \\| null`	Human-readable explanation of detected biases
`ai_section`	`string \\| null`	Combined analysis section text used in output
`pdf`	`string`	Base64-encoded PDF report. Decode with `base64.b64decode()` (Python) or `atob()` (JS).
`error`	`string \\| null`	Non-null if a non-fatal error occurred during processing

{
  "title": "Example Article",
  "content": "Full article text...",
  "source": "https://example.com/article",
  "word_count": 842,
  "summary": "A brief summary.",
  "scam_notes": null,
  "subjective_notes": "The article contains editorial language.",
  "bias_notes": "One bias type exceeded the confidence threshold.",
  "ai_section": "Combined AI analysis content.",
  "pdf": "<base64 string>",
  "error": null
}

Internal models¶

These models are used internally by the API and are not returned directly to clients, but are documented here for reference.

ExtractionResult¶

Produced by the extractor service before analysis begins.

Field	Type	Description
`title`	`string`	Extracted document title
`content`	`string`	Extracted text body
`input_type`	`string`	Classification of the input (`url`, `file`, or `text`)
`source`	`string`	Source identifier
`word_count`	`integer`	Word count of the extracted content
`extraction_method`	`string`	Method used to extract content (e.g. `newspaper4k`)
`error`	`string \\| null`	Set if extraction partially failed

ScamAnalysisResult¶

Field	Type	Description
`scam_probability`	`float` (0–1)	Confidence score that the content is a scam
`non_scam_probability`	`float` (0–1)	Confidence score that the content is legitimate
`is_scam`	`boolean`	Final classification
`notes`	`string \\| null`	Explanation
`error`	`string \\| null`	Set if analysis failed

ObjectivityAnalysisResult¶

Field	Type	Description
`subjective_probability`	`float` (0–1)	Confidence the content is subjective/opinionated
`objective_probability`	`float` (0–1)	Confidence the content is objective/factual
`is_subjective`	`boolean`	Final classification
`notes`	`string \\| null`	Explanation
`error`	`string \\| null`	Set if analysis failed

SentimentAnalysisResult¶

Field	Type	Description
`bias_cutoff`	`float` (0–1)	Minimum confidence score required for a bias label to be reported
`bias_scores`	`object`	Map of all bias category labels to their confidence scores
`biases_above_cutoff`	`string[]`	Labels whose scores met or exceeded `bias_cutoff` — surfaced as `biases` in `ClearviewResponse`
`notes`	`string \\| null`	Explanation
`error`	`string \\| null`	Set if analysis failed