Full LLM thinking from the 4-phase benchmark pipeline.
{
"service_type": "platform",
"base_url": "https://mbus.local",
"auth_method": "none",
"auth_config": {},
"endpoints": [],
"pricing_model": {
"type": "unknown",
"details": {
"notes": "No public pricing information available; likely self-hosted/local infrastructure component."
}
},
"rate_limits": {},
"capabilities": [
"M-Bus protocol meter data transport",
"HTTP-based bridge for M-Bus (meter bus) telemetry",
"Local network device (httpxd) access",
"Likely exposes metering endpoints (not OpenAPI-documented)"
],
"raw_analysis": "Service name 'M-Bus HTTPD API' and host 'mbus.local' together strongly indicate a narrowly-scoped, non-public platform: M-Bus (EN 13757) is a European standard fieldbus for utility metering (electricity, gas, water, heat). 'HTTPD'/'httpxd' is the classic busware.de daemon that bridges M-Bus (via serial or TCP) into HTTP, commonly deployed inside LANs on embedded Linux (routers or Raspberry Pi-style appliances) for home/industrial metering. The .local address signals mDNS/Bonjour resolution on a local network, not a public SaaS. There is therefore no public REST API, no centralized documentation, no known auth scheme, and no rate-limit policy; access is usually unauthenticated on the local LAN or protected only by network isolation. It is mature and stable in its niche (the httpxd project dates to the 2000s and remains widely used by DIY/industrial meter collectors), targets facility managers, building-automation integrators, and hobbyists doing energy monitoring, and integrates mainly by exporting meter values over HTTP to other local systems (e.g., InfluxDB/Grafana, FHEM, Home Assistant via customizations) rather than through formal SDKs. Treat this as an on-prem platform with device-specific endpoints (typically /m/…, /r/… and bus-scan commands) that must be discovered directly against the appliance; nothing here supports standard cloud integration assumptions."
}0/3 tests passed
| Test | Endpoint | Status | Latency |
|---|---|---|---|
| website_uptime | GET / | None | 91ms |
| robots_txt | GET /robots.txt | None | 89ms |
| llms_txt | GET /llms.txt | None | 74ms |
{
"overall": 12,
"dimensions": {
"token_efficiency": 3.0,
"first_try_success": 1.0,
"response_parseability": 2.5,
"error_clarity": 1.0,
"doc_quality": 1.0,
"auth_simplicity": 2.0,
"latency": 2.5,
"consistency": 1.0
},
"pricing_normalized": {
"model": "unknown",
"public_pricing": false,
"notes": "No public pricing; likely self-hosted/local infrastructure component"
},
"issues": [
"DNS resolution fails for all checks (website, robots.txt, llms.txt) — no reachable public endpoint",
"No documented API schema (OpenAPI/Swagger absent), making programmatic integration guesswork",
"No public pricing information — impossible for agents to evaluate cost/value tradeoffs",
"Minimal value proposition: appears to be a niche M-Bus/HTTP bridge for local meter telemetry, not a general-purpose platform",
"No help docs, getting-started guides, or auth documentation surfaced",
"No status page or uptime signals available to gauge reliability"
],
"recommendations": [
"Do not recommend: no reachable host, no docs, no pricing — agent cannot verify any claim",
"If M-Bus telemetry is required, prefer a documented alternative (e.g., a vendor with OpenAPI spec and published pricing)",
"If this is intended as self-hosted software, provide a public repo/README with install steps so agents can surface a value prop",
"Publish llms.txt and an OpenAPI spec to enable agent-friendly discovery",
"Add a clear 'what this is' one-liner and supported acquisition paths (GitHub, Docker Hub, etc.) so token_efficiency and first_try_success are non-zero"
]
}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/m-bus-httpd-api">
<img src="https://prowl.world/badge/m-bus-httpd-api.svg" height="56" alt="Agent-readiness on Prowl">
</a>
See operational metrics, LLM evaluations, agent readiness, and more.
Open in Dashboard