io.github.agenttrust-ai/agenttrust
AgentTrust — Pre-Invocation AI Agent Trust & Reliability Checks
Check an AI agent endpoint before you call it: read-only, no API key, returns a trustDecision.
alive probed 2026-10-11 · HTTP 200
1tools
347tokens in Claude Opus 5.5
357tokens in Claude Code
0.2 %of a 200k context window
201o200k_base estimate
2.5 sto initialize
257 msto list tools
- Endpoint
- https://getagenttrust.com/api/mcp/public (streamable-http)
- Host
- getagenttrust.com
- Protocol
- 2025-11-25
- Reports itself as
- agenttrust 1.0.4
- Capabilities
- tools
- Instructions
- 520 characters
- Claude count
- Claude Opus 5.5, measured 2026-10-11. Counted again when the tool definitions change. Method.
- Repository
- https://github.com/agenttrust-ai/agenttrust
- Website
- https://getagenttrust.com/docs
- Registry
- active, updated 2026-10-02 · registry entry
- Reachable
- 10 of the last 10 probes
Most expensive tools
| Tool | Tokens (o200k) | Share of this server |
|---|---|---|
| check_agent_trust | 201 | 100 % |
History
We add a row when the result changes. A server with no change keeps its last row.
| Probed | Result | Init | Tools | Claude | o200k |
|---|---|---|---|---|---|
| 2026-10-11 10:23 | alive | 2.5 s | 1 | 347 | 201 |
| 2026-10-10 10:15 | paid (402) | ||||
| 2026-10-09 11:18 | alive | 1.5 s | 1 | 347 | 201 |
| 2026-10-08 11:12 | paid (402) | ||||
| 2026-10-05 14:30 | alive | 910 ms | 1 | 347 | 201 |
| 2026-10-05 00:31 | alive | 601 ms | 1 | 201 | |
| 2026-10-04 10:03 | alive | 2.4 s | 1 | 206 | |
| 2026-10-04 07:57 | alive | 1.5 s | 1 | 206 | |
| 2026-10-03 20:36 | alive | 1.1 s | 1 | 206 | |
| 2026-10-03 17:52 | alive | 540 ms | 1 | 206 |
Badge and feed
[](https://mcpcost.ulehla.dev/s/io.github.agenttrust-ai/agenttrust)
An entry for every probe that changed the result, the tool count or the token cost: Atom feed. Add it to a feed reader or a Slack RSS app to hear when the server stops answering or its tool set grows.