Profanity Detection avatar

Profanity Detection

Pricing

$1.00 / 1,000 requests

Go to Apify Store
Profanity Detection

Profanity Detection

Analyze text content for profanity, offensive language, and inappropriate content with Machine Learning. Detects harmful language, provides risk scoring, and helps maintain safe user environments by filtering inappropriate content.

Pricing

$1.00 / 1,000 requests

Rating

0.0

(0)

Developer

Greip

Greip

Maintained by Community

Actor stats

0

Bookmarked

8

Total users

0

Monthly active users

5 days ago

Last modified

Features

Limited permissions

Share

Greip Profanity Detection (API v2)

Apify Actor that scores text for profanity risk using the Greip API v2 POST /score/text endpoint. Results include a risk level and a recommended action (accept, review, or reject) without echoing submitted text in the dataset.

Features

  • Greip API v2: JSON POST https://v2.greipapi.com/score/text with Bearer authentication
  • Risk level: low, medium, or high
  • Recommendation: accept, review, or reject
  • Secure logging: logs text length only, never raw content
  • Structured output: dataset rows with riskLevel, recommendation, and timestamp

Input parameters

ParameterTypeRequiredDefaultDescription
textstringYesText to score for profanity risk
modestringNoliveGreip mode: live (production) or test (sandbox)

Set GREIP_API_KEY in Actor environment variables. Greip v2 accepts the key only in the Authorization: Bearer header.

Output format

Successful run (dataset item):

{
"riskLevel": "low",
"recommendation": "accept",
"timestamp": "2026-11-15T13:00:00.000Z"
}

Other Greip v2 examples:

riskLevelrecommendation
mediumreview
highreject

Error dataset item (no echoed text):

{
"timestamp": "2026-11-15T13:00:00.000Z",
"error": "Greip API error 101 invalid_key",
"status": "error"
}

Output field descriptions

  • riskLevel: Overall profanity risk (low, medium, high)
  • recommendation: Suggested action from Greip (accept, review, reject)
  • timestamp: ISO 8601 time when the row was written
  • error: Present on failures; contains only Greip code and type, never Greip description
  • status: error on failed rows

BREAKING changes in 2.0

This release migrates from Greip v1 GET /scoring/profanity to v2 POST /score/text.

Removed inputs

  • listBadWords
  • scoreOnly
  • userID (v2 ScoreTextRequest and profanity docs document text, format, mode, and source only; no user_id)

Greip dashboard user attribution is no longer available.

Removed outputs

  • text
  • isML
  • riskScore
  • isSafe
  • totalBadWords
  • badWords

New or changed behavior

  • Optional input mode (live | test, default live), confirmed for /score/text in OpenAPI GreipServiceRequestFields and Greip docs
  • Request body always includes source: "Apify"
  • Dataset success fields are only riskLevel, recommendation, timestamp

Use cases

  • Content moderation for comments, posts, and messages
  • Username and display-name screening
  • Chat and community guideline enforcement

Development

npm install
npm run build
npm run lint
npm test

Feedback

Technical feedback or bugs: Apify Issues.