Tisane

{
  "service_type": "platform",
  "base_url": "https://tisane.ai",
  "auth_method": "api_key",
  "auth_config": {
    "header": "Authorization",
    "scheme": "Bearer",
    "api_key_url": "https://tisane.ai/register"
  },
  "endpoints": [],
  "pricing_model": {
    "type": "freemium",
    "details": {
      "free_tier": "Offers a free tier with limited requests for evaluation purposes",
      "paid_tiers": "Various subscription plans based on usage volume and features"
    }
  },
  "rate_limits": {
    "free_tier": "Typically lower request rate limits, e.g., 1 request per second",
    "paid_tiers": "Higher or customizable limits depending on plan"
  },
  "capabilities": [
    "abusive content detection",
    "hate speech detection",
    "threat detection",
    "offensive language detection",
    "profanity filtering",
    "content moderation",
    "text analytics",
    "sentiment analysis",
    "language detection",
    "topic classification",
    "entity recognition",
    "law enforcement text analysis",
    "multi-language support",
    "API-based integration"
  ],
  "raw_analysis": "Tisane is a text analytics platform specializing in detecting abusive content, hate speech, threats, and other harmful language. It positions itself for content moderation and law enforcement applications. The service is API-based, with authentication via API keys. Pricing appears to be freemium with a free tier for evaluation and paid subscription plans for production use. The platform is mature enough to target serious use cases like law enforcement, suggesting robust accuracy and compliance considerations. Integrations are primarily through REST API, possibly with SDKs for common languages. No public REST endpoints are documented in the provided materials, but typical usage involves sending text and receiving analysis results. The platform highlights multi-language support and various text analytics features beyond abuse detection, such as sentiment and entity recognition. For automated integration, users would need to register for an API key and review the official documentation for endpoint details."
}

2/3 tests passed

```json
{
  "overall": 68,
  "dimensions": {
    "token_efficiency": 7.5,
    "first_try_success": 5.5,
    "response_parseability": 8.0,
    "error_clarity": 5.5,
    "doc_quality": 6.5,
    "auth_simplicity": 6.0,
    "latency": 6.5,
    "consistency": 7.0
  },
  "pricing_normalized": {
    "free_tier": true,
    "paid_tiers_available": true,
    "model": "freemium"
  },
  "issues": [
    "No llms.txt file found (404) — reduces discoverability for agent-based onboarding",
    "robots.txt disallows /admin/ paths but sitemaps are also under /admin — unclear documentation availability",
    "Latency at ~630ms for homepage is acceptable but not great for agent interactions",
    "Auth process unclear — needs manual exploration to determine if it's API-key based or requires human signup"
  ],
  "recommendations": [
    "Add an llms.txt file with API endpoints, auth methods, and usage examples to improve agent onboarding",
    "Move sitemaps out of /admin/ and consider making documentation public at /docs with clear API reference",
    "Provide code samples in any agent-facing docs to reduce first-try friction",
    "Consider adding a dedicated getting-started endpoint for programmatic access"
  ]
}
```