MCP Selection Lab

io.github.ChanghuLiu/mcp-selection-labVersion 0.5.2

Benchmark MCP tool selection with metadata-only routing, collision, abstention, and holdout checks.

Status

Responds113 msNov 14, 2026, 14:36 UTCAnswered our handshake without sign-in.

Registry data may be out of date

We could not refresh it, so this page shows the last copy.

Last checked: Nov 14, 2026, 14:36 UTC
Latency: 113 ms
Remotes responding: 1 of 1
Sign-in: No sign-in
Hosting: Hosted outside the EU (detected)
Registry data as of: Nov 14, 2026, 22:14 UTC

Data from the official MCP Registry, checked by run.zero. How we check

Connect

Streamable HTTP

https://agenttrustlab.com/mcpResponds113 msNov 14, 2026, 14:36 UTC

No sign-in

Claude Code

claude mcp add --transport http mcp-selection-lab https://agenttrustlab.com/mcp

Claude Desktop and claude.ai

1
2
3

Settings > Connectors > Add custom connector
Name: mcp-selection-lab
Remote MCP server URL: https://agenttrustlab.com/mcp

Cursor

{
  "mcpServers": {
    "mcp-selection-lab": {
      "url": "https://agenttrustlab.com/mcp"
    }
  }
}

VS Code

code --add-mcp '{"name":"mcp-selection-lab","type":"http","url":"https://agenttrustlab.com/mcp"}'

Built your own server? run.zero hosts MCP servers on European infrastructure.

Host your own MCP server on run.zero

Tools

3 tools

Tools of MCP Selection Lab
	Description		Read-only	Idempotent	Open world
`get_selection_report`Read and agent-enrich an existing public scan report by id; no new target request or model call.	Read and agent-enrich an existing public scan report by id; no new target request or model call.	No	Yes	Yes	No
`scan_mcp_metadata`Create a free public report of MCP tool-description routing collisions. Reads discovery metadata (initialize/tools/list) only; never calls target business tools or spends funds. Use a public HTTP(S) URL without credentials, query or fragment. Results use at most 12 tools and 24 generated cases, not a real-model evaluation. The returned report is enriched with agent_plan candidate description-only edits and a machine-readable rerun instruction. Candidate edits are not untouched-holdout proof. Reports persist and are public by link. For an existing report use get_selection_report. Set internal_test=true for owner/CI validation so it is excluded from public scans. source is a self-reported referral bucket, not identity verification.	Create a free public report of MCP tool-description routing collisions. Reads discovery metadata (initialize/tools/list) only; never calls target business tools or spends funds. Use a public HTTP(S) URL without credentials, query or fragment. Results use at most 12 tools and 24 generated cases, not a real-model evaluation. The returned report is enriched with agent_plan candidate description-only edits and a machine-readable rerun instruction. Candidate edits are not untouched-holdout proof. Reports persist and are public by link. For an existing report use get_selection_report. Set internal_test=true for owner/CI validation so it is excluded from public scans. source is a self-reported referral bucket, not identity verification.	No	No	No	Yes
`selection_lab_info`Get free service capabilities, limits, privacy and connection details; does not scan a target.	Get free service capabilities, limits, privacy and connection details; does not scan a target.	No	Yes	Yes	No

Details

Registry status: Active
Published: Oct 13, 2026
Updated: Oct 13, 2026
Website: https://agenttrustlab.com/⁠ (opens in a new tab)
Repository: https://github.com/ChanghuLiu/mcp-selection-lab⁠ (opens in a new tab)
Categories (suggested): SecuritySuggested categories are assigned automatically from the registry text and may be wrong. Verified operators can correct them.
GDPR: Not declared

Are you the operator? Claim this listing