Auto-registered via benchmark URL submission.
Full LLM thinking from the 4-phase benchmark pipeline.
MCP server at https://db.multivoice.ai/mcp (endpoint source: mcp_manifest)
0/2 tests passed
| Test | Endpoint | Status | Latency |
|---|---|---|---|
| mcp_initialize | POST https://db.multivoice.ai/mcp | 401 | 414ms |
| mcp_tools_list | POST https://db.multivoice.ai/mcp | 401 | 145ms |
{
"overall": 58,
"dimensions": {
"auth_simplicity": 5.5,
"error_clarity": 7.0,
"doc_quality": 7.0,
"latency": 9.0
},
"pricing_normalized": {},
"issues": [
"Auth wall hit on both initialize and tools/list (401), so tool behavior, response shapes, and token efficiency could not be observed.",
"No WWW-Authenticate / JSON-RPC error body was captured in the test output, making it unclear which auth scheme (OAuth2 vs. API key Bearer) an agent should use.",
"Server was declared via mcp_manifest, so the endpoint is legitimately the operator's — the 401 confirms a live MCP server and is not a defect."
],
"recommendations": [
"Return a 401 with a WWW-Authenticate header and a JSON-RPC error body naming the required scheme so agents can self-recover.",
"If OAuth2 is used, document dynamic client registration to reduce agent onboarding friction.",
"If an API key is acceptable, allow it as a Bearer token — this is the lowest-friction path for agents.",
"Publish auth details in the MCP manifest so agents know credentials are required before probing."
]
}MCP API for agents to order MultiVoice AI audiobooks, subtitles, transcription and video dubbing: get a EUR quote, confirm with the human, then download the files.
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/multivoice-api">
<img src="https://prowl.world/badge/multivoice-api.svg" height="56" alt="Agent-readiness on Prowl">
</a>
See operational metrics, LLM evaluations, agent readiness, and more.
Open in Dashboard