POST /v1/harm/detect

Detect potentially harmful content in text.

Endpoint

POST https://api.glyphnet.io/v1/harm/detect

Authentication

Requires API key in X-API-Key header.

Request Body

Field	Type	Required	Description
`text`	string	Yes	Text to analyze (max 10,000 characters)
`categories`	array	No	Specific categories to check
`threshold`	string	No	`low`, `medium`, `high`. Default: `medium`

Available Categories

self_harm - Self-harm or suicide-related content
violence - Violent or threatening content
hate_speech - Discriminatory or hateful content
illegal - Instructions for illegal activities
explicit - Sexually explicit content
misinformation - Known false claims

Example Request

import requests
 
response = requests.post(
    "https://api.glyphnet.io/v1/harm/detect",
    headers={
        "X-API-Key": "gn_live_your_key_here",
        "Content-Type": "application/json"
    },
    json={
        "text": "Your text to analyze here...",
        "threshold": "medium"
    }
)
 
result = response.json()
if result["harmful"]:
    print(f"Harmful content detected: {result['categories']}")
else:
    print("Content is safe")

Response

Safe Content

{
  "request_id": "req_harm_abc123",
  "harmful": false,
  "categories": [],
  "confidence": 0.98,
  "processing_time_ms": 23
}

Harmful Content Detected

{
  "request_id": "req_harm_def456",
  "harmful": true,
  "categories": [
    {
      "category": "violence",
      "level": "high",
      "confidence": 0.92,
      "matched_phrases": ["..."],
      "recommendation": "This content contains violent themes and should be blocked."
    }
  ],
  "confidence": 0.92,
  "recommendation": "Block this content from being displayed.",
  "processing_time_ms": 31
}

Response Fields

Field	Type	Description
`request_id`	string	Unique request identifier
`harmful`	boolean	True if harmful content detected
`categories`	array	Detected harm categories
`confidence`	number	Overall confidence (0.0-1.0)
`recommendation`	string	Suggested action
`processing_time_ms`	number	Processing time

Category Object

Field	Type	Description
`category`	string	Category name
`level`	string	`low`, `medium`, `high`, `critical`
`confidence`	number	Confidence for this category
`matched_phrases`	array	Phrases that triggered detection
`recommendation`	string	Category-specific recommendation

Threshold Levels

Threshold	Description	Use Case
`low`	Very sensitive, catches more	User-facing content
`medium`	Balanced detection	General use
`high`	Less sensitive, fewer false positives	Internal tools

Harm Levels

Level	Description	Recommended Action
`low`	Minor concern	Log and monitor
`medium`	Moderate concern	Flag for review
`high`	Serious concern	Block content
`critical`	Immediate risk	Block and alert

Integration Example

def moderate_content(text: str) -> dict:
    """Check content before displaying to users."""
    response = requests.post(
        "https://api.glyphnet.io/v1/harm/detect",
        headers={"X-API-Key": os.environ["GLYPHNET_API_KEY"]},
        json={"text": text, "threshold": "low"}
    )
 
    result = response.json()
 
    if result["harmful"]:
        # Get the highest severity category
        max_level = max(result["categories"], key=lambda c:
            {"low": 1, "medium": 2, "high": 3, "critical": 4}[c["level"]]
        )
 
        if max_level["level"] in ["high", "critical"]:
            return {"allowed": False, "reason": result["recommendation"]}
        else:
            return {"allowed": True, "flagged": True, "warning": result["recommendation"]}
 
    return {"allowed": True, "flagged": False}

Error Responses

Same error format as /v1/verify. See Error Codes.

POST /v1/verify API Keys