MCP serverio.github.agenttrust-ai/agenttrust
Check an AI agent endpoint before you call it: read-only, no API key, returns a trustDecision.
Overview
Score?
UNRATED 0.481
of what a free look can see, on 14 looks
Looks
16
last 15 hr ago
Tools
5
changed 7 days ago
More info
URL
getagenttrust.com/api/mcp
streamable-http
Says it is
agenttrust 1.0.4
protocol 2025-06-18
In the record since
14 days ago
Among servers18,413 with a card
0median 0.606 · this server 0.481 · highest on record 0.8561
Score, from free looks onlycomputed 2 days ago
UN
RATED
RATED
0.481
14 looks · 30 days · measures answering, whether its declared schemas are well formed and speed
| Component | Weight ? | Lower bound | Adds |
|---|---|---|---|
| liveness it answered at all | 0.43 | 0.56 | 0.241 |
| schema conformance the answer matched the shape it published | 0.43 | 0.56 | 0.241 |
| latency p95 it answered as fast as its class | 0.14 | 0.00 | 0.000 |
| Composite | 1.00 | 0.481 |
Not measured: correctness, honesty, price stability ?
BURN_INFREE_TIER_ONLY
Verify it yourself
npx teppi-check https://getagenttrust.com/api/mcpcurl -s https://api.teppi.xyz/v1/trust/mcp/mcs_01M2Z1A2JB6F6BH7Y49FPS9SBB