Full LLM thinking from the 4-phase benchmark pipeline.
{
"service_type": "platform",
"base_url": "http://api.olhovivo.sptrans.com.br/v2.1",
"auth_method": "api_key",
"auth_config": {
"api_key": "",
"key_header": "Authorization",
"auth_endpoint": "/Login/Autenticar?token={api_key}"
},
"endpoints": [
{
"path": "/Login/Autenticar?token={api_key}",
"method": "POST",
"description": "Authenticate to get a session cookie"
},
{
"path": "/Posicao",
"method": "GET",
"description": "Get real-time position of all buses on a line"
},
{
"path": "/Linha/Buscar?termosBusca={term}",
"method": "GET",
"description": "Search for lines by term"
},
{
"path": "/Parada/Buscar?termosBusca={term}",
"method": "GET",
"description": "Search for stops by term"
},
{
"path": "/Previsao",
"method": "GET",
"description": "Get arrival predictions for a line/stop"
},
{
"path": "/Previsao/Linha",
"method": "GET",
"description": "Get arrival predictions for a line"
},
{
"path": "/Previsao/Parada",
"method": "GET",
"description": "Get arrival predictions for a stop"
}
],
"pricing_model": {
"type": "subscription",
"details": {
"cost": "Free with registration",
"registration_required": true,
"access_granted_via": "API key provided after registration on developer portal"
}
},
"rate_limits": {},
"capabilities": [
"Real-time bus positioning",
"Line search",
"Stop search",
"Arrival prediction",
"Public transit data for São Paulo, Brazil",
"API for developers"
],
"raw_analysis": "SPTrans operates the public transit system in São Paulo, Brazil. Their API, called 'API do Olho Vivo', provides real-time data on bus locations, line and stop information, and arrival predictions. It is a government-provided developer API, free of charge but requires registration and an API key. The API is documented in Portuguese and aimed at developers building transit applications. It is a mature platform, operational for many years, and widely used by local developers. The API uses a token-based authentication, requiring a POST to log in, then session cookies for subsequent requests. The documentation is available at the given URL, and the API is REST-like but with a focus on GET requests for data retrieval. There is no public pricing, but it is known to be free. The rate limits are not clearly documented in the public guide, so I left them empty. The base URL is the standard endpoint for the API version 2.1. The capabilities list highlights the core features of the API."
}1/3 tests passed
| Test | Endpoint | Status | Latency |
|---|---|---|---|
| website_uptime | GET / | 200 | 747ms |
| robots_txt | GET /robots.txt | 404 | 191ms |
| llms_txt | GET /llms.txt | 404 | 213ms |
```json
{
"overall": 55,
"dimensions": {
"token_efficiency": 6.0,
"first_try_success": 5.0,
"response_parseability": 6.5,
"error_clarity": 4.0,
"doc_quality": 5.0,
"auth_simplicity": 4.0,
"latency": 6.0,
"consistency": 5.5
},
"pricing_normalized": {},
"issues": [
"No robots.txt or llms.txt found — not optimized for language model discoverability",
"Documentation appears external (links to sptrans.com.br) rather than embedded guidance",
"No security headers detected on landing page",
"Registration required for API key adds onboarding friction",
"No LLM-friendly API reference paths detected in checks"
],
"recommendations": [
"Add /llms.txt and robots.txt to expose API structure to AI agents",
"Provide inline documentation snippets and self-contained getting-started guide",
"Simplify auth flow with OAuth or temporary demo keys for quick trials",
"Add structured response schemas (JSON examples) to the developer portal",
"Include rate limit and error code documentation explicitly"
]
}
```Show your live agent-readiness score on your own site. Free, no auth — it updates as your score changes.
<a href="https://prowl.world/service/transport-for-sao-paulo-brazil">
<img src="https://prowl.world/badge/transport-for-sao-paulo-brazil.svg" height="56" alt="Agent-readiness on Prowl">
</a>
See operational metrics, LLM evaluations, agent readiness, and more.
Open in Dashboard