Webpage Moderation

Submit a webpage for moderation to detect policy, safety, and compliance risks across its text, images, audio, and video content. You can submit a URL to crawl, raw page text, or raw HTML source. The service extracts every component, moderates each one, and returns a consolidated verdict.

Requirements

Item	Specification
Encoding	UTF-8
Method	POST
Recommended timeout	15s

Request

Request URL

Cluster	Request URL
USA (Virginia)	`http://api-article-fjny.fengkongcloud.com/webpage/v4`
Singapore	`http://api-article-xjp.fengkongcloud.com/webpage/v4`

Request body is JSON in the HTTP body. The body must not exceed 3.5 MB.

Request Parameters

Parameter	Type	Required	Description
`accessKey`	string	Yes	API authentication key. The default `accessKey` is sent in your onboarding email.
`appId`	string	Yes	Application identifier, such as `web` for your web application or `app` for your mobile app. The default `appId` is sent in your onboarding email. Contact DeepCleer if you need a new `appId`.
`eventId`	string	Yes	Event identifier used to distinguish moderation scenarios in your application. The default `eventId` is sent in your onboarding email. Contact DeepCleer if you need a new `eventId`.
`imgType`	string	Yes	Image detection types for images on the page. Multiple values can be joined with underscores — for example, `POLITY_QRCODE_ADVERT`. `NONE` cannot be combined with any other value. See Image Detection Types.
`txtType`	string	Yes	Text detection types for text extracted from the page. Multiple values can be joined with underscores — for example, `TEXTRISK_FRAUD`. Combined types are unioned. `NONE` cannot be combined with any other value. See Text Detection Types.
`audioType`	string	No	Audio detection types for audio on the page. Multiple values can be joined with underscores — for example, `POLITY_EROTIC_MOAN_ADVERT` (recommended). `NONE` cannot be combined with any other value. See Audio Detection Types.
`videoImgType`	string	No	Detection types for video frames sampled from page video. Multiple values can be joined with underscores. `NONE` cannot be combined with any other value. See Image Detection Types.
`videoAudioType`	string	No	Detection types for the audio track of page video. Multiple values can be joined with underscores. `NONE` cannot be combined with any other value. See Audio Detection Types.
`callback`	string	No	Callback HTTP/HTTPS URL. When provided, DeepCleer posts the full moderation result to this endpoint as soon as it becomes available.
`articleDoubleJumpConfig`	json_object	No	Second-hop crawl configuration. See `articleDoubleJumpConfig` Object.
`articleScreenShotConfig`	json_object	No	Screenshot moderation configuration. See `articleScreenShotConfig` Object.
`articleDynamicConfig`	json_object	No	Dynamic-page (rendered) moderation configuration. See `articleDynamicConfig` Object.
`data`	json_object	Yes	Request data content. Max 1 MB. See `data` Object.

Image Detection Types

Applies to imgType and videoImgType.

Value	Description
`NONE`	Do not moderate images
`POLITY`	Political content
`EROTIC`	Pornographic and sexually suggestive content
`VIOLENT`	Violence, terrorism, and prohibited content
`QRCODE`	QR code recognition
`ADVERT`	Advertising recognition
`IMGTEXTRISK`	Risky text inside images. Required if you need to moderate text rendered within an image.
`BOCR`	Minority-language OCR and automatic language detection. Singapore cluster only.

Text Detection Types

Applies to txtType.

Value	Description
`NONE`	Do not moderate text
`POLITY`	Political content
`VIOLENT`	Violence and terrorism
`BAN`	Prohibited content
`EROTIC`	Pornographic content
`DIRTY`	Profanity and abusive language
`ADVERT`	Advertising
`PRIVACY`	Personally identifiable information
`ADLAW`	Advertising law violations
`MEANINGLESS`	Meaningless or junk text
`FRAUD`	Online fraud and scams
`UNPOACH`	High-value user poaching attempts
`TEXTMINOR`	Content involving minors
`TEXTRISK`	Bundled general-risk check. Covers `POLITY`, `VIOLENT`, `BAN`, `EROTIC`, `DIRTY`, `ADVERT`, `PRIVACY`, `ADLAW`, and `MEANINGLESS`. When combined with any of these individual categories, the result is their union — so `TEXTRISK_POLITY` behaves the same as `TEXTRISK` alone.

Audio Detection Types

Applies to audioType and videoAudioType.

Value	Description
`NONE`	Do not moderate audio
`AUDIOPOLITICAL`	Voiceprint recognition of top political figures
`POLITY`	Political content
`EROTIC`	Pornographic content
`ADVERT`	Advertising
`ADLAW`	Advertising law violations
`BAN`	Prohibited content
`VIOLENT`	Violence and terrorism
`ANTHEN`	National anthem recognition
`MOAN`	Moaning / erotic breathing
`DIRTY`	Profanity and abusive language
`BANEDAUDIO`	Prohibited songs
`COPYRIGHTSONGS`	Copyrighted songs

`data` Object

Provide exactly one of url, text, or contents.

Parameter	Type	Required	Description
`url`	string	No	Downloadable URL of the webpage to moderate. The URL's `Content-Type` header must be `text/html`. Content size up to 500 MB. Text length defaults to a 500,000-character limit; images default to a 500-image limit. Provide exactly one of `url`, `text`, or `contents`.
`text`	string	No	Plain text of the page to moderate. Up to 500,000 characters. Provide exactly one of `url`, `text`, or `contents`.
`contents`	string	No	Raw HTML source of the page to moderate. Text length defaults to a 500,000-character limit; images default to a 500-image limit. Provide exactly one of `url`, `text`, or `contents`.
`lang`	string	Yes	Language of the page text. Pass `en` by default. For international traffic when the language is unknown, pass `auto` to enable automatic language detection. See Supported Languages.
`acceptLang`	string	Yes	Language for returned labels. Set `en` by default.
`returnAllImg`	int	No	Image return level. `0`: return only images whose risk level is not `PASS`. `1`: return images of all risk levels. Default `0`.
`returnAllText`	int	No	Text return level. `0`: return only text whose risk level is not `PASS`. `1`: return text of all risk levels. Default `0`.
`returnAllVideoImg`	int	No	Video-frame return level. `0`: return only frames whose risk level is not `PASS`. `1`: return all frames. Default `0`.
`returnAllVideoAudio`	int	No	Video-audio return level. `0`: return only segments whose risk level is not `PASS`. `1`: return all segments. Default `0`.
`returnAllVideo`	int	No	Video return level. `0`: return only videos whose risk level is not `PASS`. `1`: return all videos. Default `0`.
`returnAllAudio`	int	No	Audio-segment return level. `0`: return only segments whose risk level is not `PASS`. `1`: return all segments. Default `0`.
`tokenId`	string	Yes	Stable identifier for the end user, typically your internal user UID (an encrypted UID is fine). Used for behavioral-risk signals such as spam and repeat-offender detection. Alphanumeric with underscores and hyphens, up to 64 characters. Use a unique data ID if no user UID exists.
`receiveTokenId`	string	No	`tokenId` of the message recipient in a one-to-one scenario. Alphanumeric with underscores and hyphens, up to 64 characters.
`level`	int	No	User tier, used to apply different interception policies per tier. `0`: lowest (e.g., newly registered or fully inactive). `1`: low (low activity or low level). `2`: medium. `3`: high (high activity or high level). `4`: highest (e.g., paying or VIP users).
`gender`	string	No	User gender. `male` or `female`.
`nickname`	string	No	The end user's display name, moderated for risky content.
`ip`	string	No	Public IP address of the user who submitted the page. Accepts IPv4 or IPv6.
`deviceId`	string	No	Device-fingerprint identifier issued by DeepCleer.
`dataId`	string	No	Client-side identifier attached to the moderation call. DeepCleer echoes it back with the result, letting you correlate your source record with the moderation verdict.
`extra`	json_object	No	Auxiliary parameters. See `extra` Object.

`extra` Object

Parameter	Type	Required	Description
`room`	string	No	Live-room or chat-room ID. Strongly recommended when `eventId` is `groupChat`; without it, context-aware moderation cannot be applied.
`passThrough`	json	No	Arbitrary pass-through data. The value is echoed back verbatim in the response, unchanged.

`articleDoubleJumpConfig` Object

Parameter	Type	Required	Description
`isOpen`	bool	No	Whether to enable second-hop crawling. Default off.

`articleScreenShotConfig` Object

Parameter	Type	Required	Description
`isOpen`	bool	No	Whether to enable screenshot moderation. Default off.
`width`	int	No	Screenshot width. Default `1080`.
`height`	int	No	Screenshot height. Default `6480`.
`timeoutSecond`	int	No	Screenshot timeout, in seconds. Default `30`.
`userAgent`	string	No	Browser User-Agent. Default `Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.124 Safari/537.36`.

`articleDynamicConfig` Object

Parameter	Type	Required	Description
`isOpen`	bool	No	Whether to enable dynamic-page (rendered) moderation. Default off.
`userAgent`	string	No	Browser User-Agent. Default `Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.124 Safari/537.36`.
`cookie`	string	No	Cookie used when rendering the dynamic page. Default empty.

Supported Languages

Value	Language
`en`	English
`zh`	Chinese
`ar`	Arabic
`hi`	Hindi
`es`	Spanish
`fr`	French
`ru`	Russian
`pt`	Portuguese
`id`	Indonesian
`de`	German
`ja`	Japanese
`tr`	Turkish
`vi`	Vietnamese
`it`	Italian
`th`	Thai
`tl`	Filipino
`ko`	Korean
`ms`	Malay
`auto`	Automatic detection

Response

Synchronous Response

The call returns the following base response immediately. When a callback URL is provided in the request, the full result is delivered asynchronously to that callback (see Callback Response).

Parameter	Type	Required	Description
`code`	int	Yes	Response code. See Response Codes.
`message`	string	Yes	Response message corresponding to the `code`.
`requestId`	string	Yes	Unique identifier for this request. We strongly recommend logging it on your side for investigation and tuning.

Callback Response

When a callback URL is provided, the full result is posted to it asynchronously.

ℹ️
Parameters other than code, message, and requestId are only guaranteed to be returned when code is 1100.

Parameter	Type	Required	Description
`code`	int	Yes	Response code. See Response Codes.
`message`	string	Yes	Response message corresponding to the `code`.
`requestId`	string	Yes	Unique identifier for this request. We strongly recommend logging it on your side for investigation and tuning.
`riskLevel`	string	Yes	Overall disposition recommendation. `PASS` — clean; allow it. `REVIEW` — suspicious; route to human review. `REJECT` — violates policy; block it.
`textDetails`	json_array	Yes	Risk detail for text on the page. See `textDetails` Array.
`imgDetails`	json_array	Yes	Risk detail for images on the page. See `imgDetails` Array.
`audioDetails`	json_array	Yes	Risk detail for audio on the page. See `audioDetails` Array.
`videoDetails`	json_array	Yes	Risk detail for video on the page. See `videoDetails` Array.
`auxInfo`	json_object	Yes	Auxiliary information. See `auxInfo` Object.
`resultType`	int	Yes	Current result stage. `0`: machine moderation. `1`: human moderation.
`finalResult`	int	Yes	Whether this is the final result. `1`: final (can be used directly for downstream processing; default when only machine moderation is enabled). `0`: intermediate (human review will follow).

Response Codes

Code	Message
`1100`	Success
`1901`	QPS limit exceeded
`1902`	Invalid parameters
`1903`	Service failure
`1905`	Character count exceeded
`9100`	Insufficient balance
`9101`	Unauthorized operation

`textDetails` Array

Each element describes one moderated text segment from the page.

Parameter	Type	Required	Description
`code`	int	Yes	Response code for this segment. See Response Codes.
`message`	string	Yes	Response message corresponding to the `code`.
`requestId`	string	Yes	Unique identifier for this request.
`riskLevel`	string	Yes	Disposition recommendation. `PASS`, `REVIEW`, or `REJECT`.
`riskLabel1`	string	Yes	Level-1 risk label. Returns `normal` when `riskLevel` is `PASS`.
`riskLabel2`	string	Yes	Level-2 risk label. Empty when `riskLevel` is `PASS`.
`riskLabel3`	string	Yes	Level-3 risk label. Empty when `riskLevel` is `PASS`.
`riskDescription`	string	Yes	Human-readable summary of the detected risk. Returns `Normal` when `riskLevel` is `PASS`.
`riskDetail`	json_object	Yes	Risk detail. See `riskDetail` Object.
`tokenLabels`	json_object	Yes	Account risk profile label information.
`tokenLabels.UGC_account_risk`	json_object	No	UGC content risk.
`tokenLabels.UGC_account_risk.sexy_risk_tokenid`	float	No	Pornographic account risk score (0–1).
`auxInfo`	json_object	Yes	Auxiliary information. See `textDetails.auxInfo` Object.
`allLabels`	json_array	Yes	All matched risk labels and details. See `allLabels` Array.
`businessLabels`	json_array	Yes	All matched business labels and details. See `businessLabels` Array.
`tokenProfileLabels`	json_array	No	Account attribute labels. Returned only when `tokenId` is provided and the label service is enabled. See Token Labels.
`tokenRiskLabels`	json_array	No	Account risk labels. Returned only when `tokenId` is provided and the label service is enabled. See Token Labels.
`langResult`	json_object	No	Language information. See `langResult` Object.

`riskDetail` Object

Used by textDetails.riskDetail and by every nested riskDetail of the same shape (for example, allLabels[].riskDetail).

Parameter	Type	Required	Description
`matchedLists`	json_array	No	Custom keyword lists that were triggered. Each entry names a list configured in the DeepCleer console and enumerates the matched terms. Returned only when a configured keyword appears.
`matchedLists[].name`	string	No	Name of the triggered list, as configured in the DeepCleer console.
`matchedLists[].words`	json_array	No	Matched terms from this list.
`matchedLists[].words[].word`	string	No	The matched term, as it appeared on the list.
`matchedLists[].words[].position`	int_array	No	Character offsets of the match in the input text.
`riskSegments`	json_array	No	Contiguous high-risk spans of the input. Returned when political, terrorism, prohibited, or advertising-law content is detected.
`riskSegments[].segment`	string	No	The high-risk span, extracted verbatim from the input.
`riskSegments[].position`	int_array	No	Character offsets of the span in the input text.

`textDetails.auxInfo` Object

Parameter	Type	Required	Description
`filteredText`	string	No	Text with risky segments replaced by asterisks (`*`).
`passThrough`	json_object	No	Pass-through field. Same value as `data.extra.passThrough` in the request.
`contactResult`	json_array	No	Contact information detection results, including WeChat, QQ, and phone numbers.
`contactResult[].contactType`	int	No	Contact type. `0`: phone number. `1`: QQ number. `2`: WeChat ID.
`contactResult[].contactString`	string	No	Detected contact information string.
`contextText`	string	No	Returned when context recognition is enabled.
`unauthorizedType`	string	No	Moderation types that are not enabled. Contact DeepCleer to enable them.

`allLabels` Array

Each element describes one matched risk label. Used wherever an allLabels array appears in the response.

Parameter	Type	Required	Description
`riskLevel`	string	Yes	Risk level: `REVIEW` or `REJECT`.
`riskLabel1`	string	Yes	Level-1 risk label. Required when `allLabels` is not empty.
`riskLabel2`	string	Yes	Level-2 risk label. Required when `allLabels` is not empty.
`riskLabel3`	string	Yes	Level-3 risk label. Required when `allLabels` is not empty.
`riskDescription`	string	Yes	Risk description. Required when `allLabels` is not empty. For reference only; do not build logic on this value.
`probability`	float	Yes	Confidence score (0–1). Higher values indicate greater confidence. Required when `allLabels` is not empty.
`riskDetail`	json_object	Yes	Mapped risk detail, same structure as `riskDetail` Object. Required when `allLabels` is not empty.

`businessLabels` Array

Each element describes one matched business label.

Parameter	Type	Required	Description
`businessLabel1`	string	Yes	Level-1 business label.
`businessLabel2`	string	Yes	Level-2 business label.
`businessLabel3`	string	Yes	Level-3 business label.
`businessDescription`	string	Yes	Business label description.
`probability`	float	Yes	Confidence score (0–1).
`businessDetail`	json_object	Yes	Business detail information.

`imgDetails` Array

Each element describes one moderated image from the page.

Parameter	Type	Required	Description
`requestId`	string	Yes	Unique identifier for this request.
`message`	string	Yes	Response message corresponding to the `code`.
`code`	int	Yes	Response code. `1100` Success, `1901` QPS limit exceeded, `1902` Invalid parameters, `1903` Service failure, `1911` Image download failed.
`riskLevel`	string	Yes	Disposition recommendation. `PASS`, `REVIEW`, or `REJECT`.
`riskLabel1`	string	Yes	Level-1 risk label. Returns `normal` when `riskLevel` is `PASS`.
`riskLabel2`	string	Yes	Level-2 risk label. Empty when `riskLevel` is `PASS`.
`riskLabel3`	string	Yes	Level-3 risk label. Empty when `riskLevel` is `PASS`.
`riskDescription`	string	Yes	Human-readable summary. Returns `Normal` when `riskLevel` is `PASS`.
`resultType`	int	Yes	Current result stage. `0`: machine moderation. `1`: human moderation.
`finalResult`	int	Yes	Whether this is the final result. `0`: intermediate. `1`: final.
`auxInfo`	json_object	Yes	Auxiliary information. See `imgDetails.auxInfo` Object.
`allLabels`	json_array	Yes	All matched risk labels. Each element has the same structure as `allLabels` Array, but its `riskDetail` follows Image `riskDetail` Object.
`businessLabels`	json_array	No	All matched business labels. See Image `businessLabels` Array.
`riskDetail`	json_object	Yes	Risk detail. See Image `riskDetail` Object.
`disposal`	json_object	No	Mapped disposition result. Returned only when a custom label/disposition mapping is configured. See `disposal` Object.
`tokenLabels`	json_object	Yes	Account risk profile label information. Returned only when `tokenId` is provided and the feature is enabled. See `imgDetails.tokenLabels` Object.
`tokenProfileLabels`	json_array	No	Account attribute labels. Returned only when `tokenId` is provided and the label service is enabled. See Token Labels.
`tokenRiskLabels`	json_array	No	Account risk labels. Returned only when `tokenId` is provided and the label service is enabled. See Token Labels.

Image `riskDetail` Object

Used by imgDetails.riskDetail, imgDetails.allLabels[].riskDetail, and disposal.riskDetail (identical structure throughout).

Parameter	Type	Required	Description
`riskSource`	int	Yes	Risk source. `1000`: no risk. `1001`: text risk. `1002`: visual image risk.
`face_num`	int	No	Number of faces detected.
`person_num`	int	No	Number of human figures detected.
`faces`	json_array	No	Recognized public figures and their positions. Up to 10 entries (highest `probability` kept if more than 10).
`faces[].face_ratio`	float	No	Face area ratio (0–1). Higher means a larger face.
`faces[].id`	string	No	Person ID. The same position across labels shares the same ID; a person appearing n times gets n IDs.
`faces[].name`	string	No	Recognized public-figure name.
`faces[].probability`	float	No	Confidence score (0–1).
`faces[].location`	int_array	No	Bounding box `[x1, y1, x2, y2]`: top-left (x1, y1) and bottom-right (x2, y2).
`objects`	json_array	No	Detected logos / objects and their positions.
`objects[].id`	string	No	Object/logo ID. The same position across labels shares the same ID.
`objects[].name`	string	No	Object name.
`objects[].probability`	float	No	Confidence score (0–1).
`objects[].qrContent`	string	No	QR code content. Returned only when a QR-code label is matched.
`objects[].location`	int_array	No	Bounding box `[x1, y1, x2, y2]`.
`ocrText`	json_object	No	OCR text recognized in the image. Present when `imgType` includes `IMGTEXTRISK` or `ADVERT`.
`ocrText.text`	string	No	Text recognized in the image.
`ocrText.matchedLists`	json_array	No	Triggered custom keyword lists. Returned only when a configured keyword appears.
`ocrText.matchedLists[].name`	string	No	Name of the triggered list.
`ocrText.matchedLists[].words`	json_array	No	Matched terms from this list.
`ocrText.matchedLists[].words[].word`	string	No	The matched term.
`ocrText.matchedLists[].words[].position`	int_array	No	Character offsets of the match.
`ocrText.riskSegments`	json_array	No	High-risk spans in the OCR text.
`ocrText.riskSegments[].segment`	string	No	The high-risk span.
`ocrText.riskSegments[].position`	int_array	No	Character offsets of the span (zero-indexed).
`persons`	json_array	No	Human figures and their positions. Up to 10 entries.
`persons[].id`	string	No	Person ID. The same person across labels shares the same ID; a person appearing n times gets n IDs.
`persons[].person_ratio`	float	No	Human-figure area ratio (0–1).
`persons[].probability`	float	No	Confidence score (0–1).
`persons[].location`	int_array	No	Bounding box of the human figure.

`imgDetails.auxInfo` Object

Parameter	Type	Required	Description
`segments`	int	Yes	Number of segments actually processed.
`typeVersion`	json_object	Yes	Model version per detection type. Format `X.Y` where `X` is the major version (overall model iteration) and `Y` is the minor version (routine iteration). Keys: `ADVERT`, `EROTIC`, `IMGTEXTRISK`, `POLITY`, `QRCODE`, `VIOLENT`.
`errorCode`	int	No	Error code. `2001`: invalid JSON. `2002`: invalid parameter field. `2003`: image download failed. `2004`: image too large (>10 MB). `2005`: invalid image format. `2006`: invalid detection type.
`frameTime`	int	No	Frame timestamp (13-digit millisecond Unix timestamp). Returned when the `streamInfo` feature is used.
`qrContent`	string	No	QR code content recognized in the image.
`streamId`	string	No	Stream identifier. Returned when stream-related parameters are passed.
`passThrough`	json_object	No	Pass-through field. Same value as `data.extra.passThrough` in the request.

Image `businessLabels` Array

Parameter	Type	Required	Description
`businessDescription`	string	Yes	Business label description, formatted as "Level1:Level2:Level3".
`businessLabel1`	string	Yes	Level-1 business label.
`businessLabel2`	string	Yes	Level-2 business label.
`businessLabel3`	string	Yes	Level-3 business label.
`probability`	float	Yes	Confidence score (0–1).
`confidenceLevel`	int	No	Confidence tier (0–2). Higher means greater confidence.
`businessDetail`	json_object	No	Business detail. Same structure as Image `riskDetail` Object, including `faces`, `objects`, `persons`, plus `face_compare_num` (face count from `imgCompareBase` when `FACECOMPARE` is used).

`disposal` Object

Returned only when a custom label/disposition mapping is configured.

Parameter	Type	Required	Description
`riskDescription`	string	Yes	Mapped risk description. `Normal` when `riskLevel` is `PASS`.
`riskLabel1`	string	Yes	Mapped level-1 risk label. `normal` when no custom label is mapped and the mapped `riskLevel` is `PASS`.
`riskLabel2`	string	Yes	Mapped level-2 risk label. Empty when no custom label is mapped and the mapped `riskLevel` is `PASS`.
`riskLabel3`	string	Yes	Mapped level-3 risk label. Empty when no custom label is mapped and the mapped `riskLevel` is `PASS`.
`riskLevel`	string	Yes	Mapped disposition recommendation. Returns the default recommendation when no custom rule matched.
`riskDetail`	json_object	Yes	Mapped risk detail, same structure as Image `riskDetail` Object.

`imgDetails.tokenLabels` Object

Parameter	Type	Required	Description
`UGC_account_risk`	json_object	No	UGC content risk.
`UGC_account_risk.b_advertise_risk_tokenid`	int	No	Advertising risk. `0`: none detected. `1`: advertising risk present.
`UGC_account_risk.b_advertise_risk_tokenid_last_ts`	int	No	Advertising risk timestamp.
`UGC_account_risk.b_politics_risk_tokenid`	int	No	Political risk. `0`: none detected. `1`: political risk present.
`UGC_account_risk.b_politics_risk_tokenid_last_ts`	int	No	Political risk timestamp.
`UGC_account_risk.b_sexy_risk_tokenid`	int	No	Pornographic risk. `0`: none detected. `1`: pornographic risk present.
`UGC_account_risk.b_sexy_risk_tokenid_last_ts`	int	No	Pornographic risk timestamp.
`machine_account_risk`	json_object	No	Bot-control risk.
`machine_account_risk.b_machine_control_tokenid`	int	No	Bot account. `0`: not bot-controlled. `1`: bot-controlled.
`machine_account_risk.b_machine_control_tokenid_last_ts`	int	No	Bot account timestamp.
`machine_account_risk.b_offer_wall_tokenid`	int	No	Offerwall account. `0`: not an offerwall account. `1`: offerwall.
`machine_account_risk.b_offer_wall_tokenid_last_ts`	int	No	Offerwall account timestamp.
`scene_account_risk`	json_object	No	Scene-specific account risk (only available in special scenarios such as airlines).
`scene_account_risk.i_tout_risk_tokenid`	int	No	Airline seat-blocking account. `0`: no. `1`: yes.
`scene_account_risk.i_tout_risk_tokenid_last_ts`	int	No	Airline seat-blocking timestamp.

`audioDetails` Array

Each element describes one moderated audio item from the page.

Parameter	Type	Required	Description
`requestId`	string	Yes	Unique identifier for this request.
`code`	int	Yes	Response code. `1100` Success, `1101` Processing, `1901` QPS limit exceeded, `1902` Invalid parameters, `1903` Service failure, `1904` Download failed, `1905` Decode failed, `9100` Insufficient balance. Fields other than `message`/`requestId` are present only when `code` is `1100`.
`message`	string	Yes	Response message corresponding to the `code`.
`riskLevel`	string	Yes	Disposition recommendation. `PASS`, `REVIEW`, or `REJECT`.
`audioText`	string	Yes	Full audio transcription.
`audioTime`	int	Yes	Total audio duration, in seconds.
`audioDetail`	json_array	Yes	Audio segment details. See `audioDetail` Array.

`audioDetail` Array

Parameter	Type	Required	Description
`requestId`	string	Yes	Unique identifier for this audio segment.
`audioStarttime`	float	Yes	Segment start time relative to the audio start, in seconds.
`audioEndtime`	float	Yes	Segment end time relative to the audio start, in seconds.
`audioUrl`	string	Yes	Segment URL (MP3).
`riskLevel`	string	Yes	Disposition recommendation. `PASS`, `REVIEW`, or `REJECT`.
`businessLabels`	json_array	No	Matched business labels. Same structure as `businessLabels` Array, plus `confidenceLevel` (0–2).
`allLabels`	json_array	No	Matched risk labels. Same structure as `allLabels` Array; `riskDetail` follows Audio `riskDetail` Object.
`riskLabel1`	string	Yes	Level-1 risk label. `normal` when `riskLevel` is `PASS`.
`riskLabel2`	string	Yes	Level-2 risk label. Empty when `riskLevel` is `PASS`.
`riskLabel3`	string	Yes	Level-3 risk label. Empty when `riskLevel` is `PASS`.
`riskDescription`	string	Yes	Human-readable summary. `Normal` when `riskLevel` is `PASS`.
`riskDetail`	json_object	No	Risk detail. See Audio `riskDetail` Object.

Audio `riskDetail` Object

Used by audioDetail.riskDetail and audioDetail.allLabels[].riskDetail.

Parameter	Type	Required	Description
`riskSource`	int	Yes	Risk source. `1000`: no risk. `1001`: text risk. `1003`: audio speech risk.
`audioText`	string	No	Audio transcription result.
`matchedLists`	json_array	No	Triggered custom keyword lists. Returned only when a configured keyword appears.
`matchedLists[].name`	string	No	Name of the triggered list.
`matchedLists[].words`	json_array	No	Matched terms from this list.
`matchedLists[].words[].word`	string	No	The matched term.
`matchedLists[].words[].position`	int_array	No	Character offsets of the match.
`riskSegments`	json_array	No	High-risk spans.
`riskSegments[].segment`	string	No	The high-risk span.
`riskSegments[].position`	int_array	No	Character offsets of the span.

`videoDetails` Array

Each element describes one moderated video from the page.

Parameter	Type	Required	Description
`code`	int	Yes	Response code. `1100` Success, `1901` QPS limit exceeded, `1902` Invalid parameters, `1903` Service failure, `1905` Character count exceeded.
`message`	string	Yes	Response message corresponding to the `code`.
`requestId`	string	Yes	Unique identifier for this request.
`riskLevel`	string	Yes	Disposition recommendation. `PASS`, `REVIEW`, or `REJECT`.
`frameDetail`	json_array	No	Frame risk detail. Returned when there is a risky frame or `returnAllVideoImg=1`. See `frameDetail` Array.
`audioDetail`	json_array	No	Video-audio segment detail. Returned when there is a risky segment or `returnAllVideoAudio=1`. See `videoDetails.audioDetail` Array.
`auxInfo`	json_object	No	Auxiliary information. Present when `code` is `1100`. See `videoDetails.auxInfo` Object.
`tokenProfileLabels`	json_array	No	Account attribute labels. Returned only when the feature is enabled. See Token Labels.
`tokenRiskLabels`	json_array	No	Account risk labels. Returned only when the feature is enabled. See Token Labels.

`frameDetail` Array

Parameter	Type	Required	Description
`requestId`	string	Yes	Unique identifier for this request.
`imgUrl`	string	Yes	Sampled frame image URL.
`riskDescription`	string	Yes	Risk description. `Normal` when `riskLevel` is `PASS`. Formatted "Level1:Level2:Level3"; for custom-list matches returns "Custom list matched".
`riskLabel1`	string	Yes	Level-1 risk label. `normal` when `riskLevel` is `PASS`.
`riskLabel2`	string	Yes	Level-2 risk label. Empty when `riskLevel` is `PASS`.
`riskLabel3`	string	Yes	Level-3 risk label. Empty when `riskLevel` is `PASS`.
`riskLevel`	string	Yes	Disposition recommendation. `PASS`, `REVIEW`, or `REJECT`.
`allLabels`	json_array	Yes	Matched risk labels. Same structure as `allLabels` Array; `riskDetail` follows Image `riskDetail` Object.
`auxInfo`	json_object	Yes	Auxiliary information. See below.
`auxInfo.similarity`	float	Yes	Similarity (0–1) between this frame and the previous frame. The first frame is compared against a pure-black image.
`auxInfo.qrContent`	string	No	QR code content recognized in the frame.
`riskDetail`	json_object	Yes	Risk detail. Same structure as Image `riskDetail` Object.
`imgText`	string	No	OCR text of the frame. Returned only when `ADVERT` or `IMGTEXTRISK` is passed.
`time`	float	No	Frame time within the video file, in seconds.
`businessLabels`	json_array	No	Matched business labels. Same structure as Image `businessLabels` Array.

`videoDetails.audioDetail` Array

Parameter	Type	Required	Description
`requestId`	string	Yes	Unique identifier for this request.
`audioUrl`	string	Yes	Audio segment URL.
`riskDescription`	string	Yes	Risk description. `Normal` when `riskLevel` is `PASS`. Formatted "Level1:Level2:Level3"; for custom-list matches returns "Custom list matched".
`riskLabel1`	string	Yes	Level-1 risk label. `normal` when `riskLevel` is `PASS`.
`riskLabel2`	string	Yes	Level-2 risk label. Empty when `riskLevel` is `PASS`.
`riskLabel3`	string	Yes	Level-3 risk label. Empty when `riskLevel` is `PASS`.
`riskLevel`	string	Yes	Disposition recommendation. `PASS`, `REVIEW`, or `REJECT`.
`allLabels`	json_array	Yes	Matched risk labels. Same structure as `allLabels` Array; `riskDetail` follows Audio `riskDetail` Object.
`audioEndtime`	float	No	Segment end time relative to the audio start, in seconds.
`audioStarttime`	float	No	Segment start time relative to the audio start, in seconds.
`audioText`	string	No	Recognized text of this segment.
`businessLabels`	json_array	No	Matched business labels. Same structure as `businessLabels` Array, plus `confidenceLevel` (0–2).
`riskDetail`	json_object	No	Risk detail. Same structure as Audio `riskDetail` Object.

`videoDetails.auxInfo` Object

Parameter	Type	Required	Description
`audioDuration`	float	Yes	Audio duration of the video in this request, in seconds (billing-aligned). If there is no audio track, this is `0`.
`frameCount`	int	Yes	Number of sampled frames returned. Risk count when `returnAllImg=0`; full count when `returnAllImg=1`.
`videoDuration`	float	Yes	Video duration, in seconds.

`auxInfo` Object

Top-level auxiliary information for the whole page.

Parameter	Type	Required	Description
`textNum`	int	Yes	Number of characters in this request (billing-aligned). Includes Chinese characters, letters, punctuation, and spaces.
`imgNum`	int	Yes	Number of images in this request (billing-aligned). Animated images are sampled into 3 frames; long images are split.
`audioNum`	int	Yes	Number of audio items in this request (billing-aligned).
`videoNum`	int	Yes	Number of videos in this request (billing-aligned).
`passThrough`	json_object	No	Pass-through field. Same value as `data.extra.passThrough` in the request.

Token Labels

Both tokenProfileLabels and tokenRiskLabels share the same structure.

Parameter	Type	Required	Description
`label1`	string	No	Level-1 label.
`label2`	string	No	Level-2 label.
`label3`	string	No	Level-3 label.
`description`	string	No	Label description.
`timestamp`	int	No	Label timestamp. 13-digit Unix timestamp in milliseconds.

`langResult` Object

Parameter	Type	Required	Description
`detectedLang`	string	No	Language detection result. Returned when `lang` is set to `auto`. Value is a standard language code (e.g., `en`, `ar`).
`translatedText`	string	No	Translation result. Returned when `translationTargetLang` is provided.

Examples

Request Example

{
  "accessKey": "YOUR_ACCESS_KEY",
  "appId": "default",
  "eventId": "webpage",
  "imgType": "POLITY_EROTIC_ADVERT",
  "txtType": "TEXTRISK",
  "audioType": "POLITY_EROTIC_ADVERT",
  "videoImgType": "POLITY_EROTIC_ADVERT",
  "videoAudioType": "POLITY_EROTIC_ADVERT",
  "callback": "http://www.example.com/callback",
  "articleDoubleJumpConfig": {
    "isOpen": true
  },
  "articleScreenShotConfig": {
    "isOpen": true,
    "width": 1080,
    "height": 6480,
    "timeoutSecond": 30,
    "userAgent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.124 Safari/537.36"
  },
  "articleDynamicConfig": {
    "isOpen": true,
    "userAgent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.124 Safari/537.36",
    "cookie": ""
  },
  "data": {
    "url": "https://www.example.com/page",
    "tokenId": "user_123456",
    "lang": "en",
    "acceptLang": "en",
    "returnAllImg": 0,
    "returnAllText": 0,
    "returnAllVideoImg": 0,
    "returnAllVideoAudio": 0,
    "returnAllVideo": 0,
    "returnAllAudio": 0,
    "level": 2,
    "gender": "male",
    "nickname": "test_user",
    "ip": "123.171.34.3",
    "deviceId": "device_fingerprint_123",
    "dataId": "data_unique_id_001",
    "extra": {
      "room": "room_001",
      "passThrough": {
        "customField": "customValue"
      }
    }
  }
}

Synchronous Response Example

{
  "code": 1100,
  "message": "success",
  "requestId": "xxx"
}

Callback Response Example

{
  "code": 1100,
  "message": "success",
  "requestId": "xxx",
  "riskLevel": "REVIEW",
  "textDetails": [
    {
      "allLabels": [
        {
          "probability": 1,
          "riskDescription": "Politics:Politics:Politics",
          "riskDetail": {
            "matchedLists": [
              {
                "name": "test_list",
                "words": [
                  {
                    "position": [10, 11],
                    "word": "58"
                  }
                ]
              }
            ]
          },
          "riskLabel1": "politics",
          "riskLabel2": "politics",
          "riskLabel3": "politics",
          "riskLevel": "REVIEW"
        }
      ],
      "requestId": "xxx_1",
      "riskDescription": "Politics:Politics:Politics",
      "riskDetail": {
        "matchedLists": [
          {
            "name": "test_list",
            "words": [
              {
                "position": [10, 11, 12],
                "word": "585"
              }
            ]
          }
        ]
      },
      "riskLabel1": "politics",
      "riskLabel2": "politics",
      "riskLabel3": "politics"
    }
  ],
  "imgDetails": [],
  "auxInfo": {
    "imgNum": 4,
    "textNum": 317
  }
}

Result Query API

Use the result query API to actively poll for the final verdict by requestId — for asynchronous queues, callback delivery, or segmented audio/video callbacks. The business fields match the Callback Response.

Request URL

Cluster	Request URL
USA (Virginia)	`http://api-article-fjny.fengkongcloud.com/query_webpage/v4`
Singapore	`http://api-article-xjp.fengkongcloud.com/query_webpage/v4`

Request body is JSON in the HTTP body. The body must not exceed 1 MB.

Request Parameters

Parameter	Type	Required	Description
`accessKey`	string	Yes	API authentication key, same as the moderation API. Provided by DeepCleer.
`requestIds`	string_array	Yes	List of detection `requestId` values returned by the webpage moderation API's synchronous response. Up to 20 per call; each up to 128 characters; duplicates are de-duplicated.

Response Parameters

Parameter	Type	Required	Description
`code`	int	Yes	Response code. `1100`: success. `1902`: the whole request is invalid. `1101` and `1903` do not appear here — they only appear in `machineResult.code`.
`message`	string	Yes	Response message corresponding to the `code`.
`requestId`	string	Yes	Identifier of this query request.
`contents`	json_array	Yes	Query results, in the same order as the requested `requestIds`. See below.

`contents` Array

Parameter	Type	Required	Description
`requestId`	string	Yes	The detection `requestId` this entry corresponds to.
`machineResult`	json_object	Yes	Machine moderation result. Per-entry `code` (independent of the top-level `code`): `1100` final success (business fields as in Callback Response); `1101` no history or still processing (message "Request is processing"); `1902` history `data` cannot be parsed as valid JSON; `1903` result written as a degraded record because the data was too long (message "Service failure", includes `detail`).
`humanResult`	json_object	No	Human moderation result. Returned when human review occurred.
`mergeResult`	json_object	No	Merged result. Shows the strategy-merged result first, e.g., `riskLevel`.

Examples

Request Example

{
  "accessKey": "YOUR_ACCESS_KEY",
  "requestIds": ["xxx_detection_request_id"]
}

Response Example — Processing (`machineResult.code` is `1101`)

{
  "code": 1100,
  "message": "Normal",
  "requestId": "query_req_001",
  "contents": [
    {
      "requestId": "xxx_detection_request_id",
      "machineResult": {
        "code": 1101,
        "message": "Request is processing",
        "requestId": "xxx_detection_request_id"
      },
      "mergeResult": {
        "riskLevel": "xxx"
      }
    }
  ]
}

Response Example — Final Success (`machineResult.code` is `1100`)

{
  "code": 1100,
  "message": "Normal",
  "requestId": "query_req_001",
  "contents": [
    {
      "requestId": "xxx_detection_request_id",
      "machineResult": {
        "code": 1100,
        "message": "Success",
        "requestId": "xxx_detection_request_id",
        "riskLevel": "xxx",
        "textDetails": [],
        "imgDetails": [],
        "audioDetails": [],
        "videoDetails": [],
        "auxInfo": {},
        "resultType": 0,
        "finalResult": 1
      },
      "mergeResult": {
        "riskLevel": "xxx"
      }
    }
  ]
}

Requirements

Request

Request URL

Request Parameters

Image Detection Types

Text Detection Types

Audio Detection Types

data Object

extra Object

articleDoubleJumpConfig Object

articleScreenShotConfig Object

articleDynamicConfig Object

Supported Languages

Response

Synchronous Response

Callback Response

Parameters other than code, message, and requestId are only guaranteed to be returned when code is 1100.

Response Codes

textDetails Array

riskDetail Object

textDetails.auxInfo Object

allLabels Array

businessLabels Array

imgDetails Array

Image riskDetail Object

imgDetails.auxInfo Object

Image businessLabels Array

disposal Object

imgDetails.tokenLabels Object

audioDetails Array

audioDetail Array

Audio riskDetail Object

videoDetails Array

frameDetail Array

videoDetails.audioDetail Array

videoDetails.auxInfo Object

auxInfo Object

Token Labels

langResult Object

Examples

Request Example

Synchronous Response Example

Callback Response Example

Result Query API

Request URL

Request Parameters

Response Parameters

contents Array

Examples

Request Example

Response Example — Processing (machineResult.code is 1101)

Response Example — Final Success (machineResult.code is 1100)

`data` Object

`extra` Object

`articleDoubleJumpConfig` Object

`articleScreenShotConfig` Object

`articleDynamicConfig` Object

Parameters other than `code`, `message`, and `requestId` are only guaranteed to be returned when `code` is `1100`.

`textDetails` Array

`riskDetail` Object

`textDetails.auxInfo` Object

`allLabels` Array

`businessLabels` Array

`imgDetails` Array

Image `riskDetail` Object

`imgDetails.auxInfo` Object

Image `businessLabels` Array

`disposal` Object

`imgDetails.tokenLabels` Object

`audioDetails` Array

`audioDetail` Array

Audio `riskDetail` Object

`videoDetails` Array

`frameDetail` Array

`videoDetails.audioDetail` Array

`videoDetails.auxInfo` Object

`auxInfo` Object

`langResult` Object

`contents` Array

Response Example — Processing (`machineResult.code` is `1101`)

Response Example — Final Success (`machineResult.code` is `1100`)