Agent Trust Lab

AIエージェントがMCPツールをどの程度確実に選択するかについての独立した証拠。凍結されたベンチマーク結果、保存された意味的ミス、マルチバッチ信頼性の証拠、および独立したMCP評価を含む。

ドキュメント

Trust profiles forAgents & MCP Servers.

Agent Trust Lab turns public artifacts, reproducible benchmark results and verified operational evidence into version-bound Trust Profiles. Selection evidence is published as a separate badge class; it never silently changes transport evidence, ownership or static profile grade.

Evaluate your MCP →

Independent MCP Selection Evaluation · US$249 introductory price.
Receive a concise evidence report with tested routing scenarios, observed findings and recommendations. Payment is requested after qualification.

36 Products

27 Rich profiles

23 Evidence tested

3 Browse categories

Evidence coverage is not a quality score. Missing evidence is shown as NOT_PROVIDED, never converted to zero. Selection badges use a separate S0–S4 evidence policy and do not improve static E0–E4 evidence grade or ranking.

Independent MCP Selection Evaluation

For MCP server developers and teams responsible for tool selection: submit a public endpoint or repository, receive qualification review, then get an independent selection evaluation and concise evidence report if accepted.

What happens next: submit details → qualification review → payment requested after acceptance → measurement → report delivery.

See the evaluation and sample report →

Product & Evidence Catalog

Evidence Tested

Browse Evidence Tested →

MCP Servers

Browse MCP Servers →

Agent Tools

Browse Agent Tools →

Developer Infrastructure

Browse Developer Infrastructure →

Recently Added

Browse Recently Added →

Independent MCP Selection Evaluation — US$249 →

Machine-readable by default

Open the catalog API →

What a Trust Profile means

Read the methodology →