Skip to content

Trust Live

by ChanghuLiu · remote · no sign-in to list tools · v0.1.0 · published by GitHub user ChanghuLiu

Evidence-linked MCP selection and non-executing Safe-Call review; surfaces unknowns.

Changes · 2
CompareRegistry entry

5 tools: 2 change things and 3 only read.

Scanned Oct 21, 2026

1 check matched something in the tool definitions. No finding is not a safety guarantee. How the checks work.

claude mcp add --transport http trust-live https://mcp-selection-lab-production.up.railway.app/trust-live/mcp

Findings

Matches of fixed checks, with the text they matched.

1 warn
  • Warnevaluate_safe_call

    The server marks "evaluate_safe_call" read-only (readOnlyHint: true), and its own name, description or input give the class "write".

    Matched text

    annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":false}; The name "evaluate_safe_call" says "evaluate": a word for a change.

    What this check looks for

Tools

Open a tool for its reasons and full description. Descriptions are the publisher's own words, shown as they are.

5 tools

Changes things 2Claude Code: ask
  • evaluate_safe_callEvaluate Safe-Call constraints without invoking the provider tool.1 finding

    Why it is in this group

    • The name "evaluate_safe_call" says "evaluate": a word for a change.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.

    Findings

    • Warn

      The server marks "evaluate_safe_call" read-only (readOnlyHint: true), and its own name, description or input give the class "write".

      Matched text

      annotations: {"readOnlyHint":true,"destructiveHint":false,"idempotentHint":true,"openWorldHint":false}; The name "evaluate_safe_call" says "evaluate": a word for a change.

      What this check looks for
    Description, as the server wrote it
    Evaluate Safe-Call constraints without invoking the provider tool.

    Annotations the server declares

    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
    • readOnlyHint: true
  • recommend_providerRecommend a provider with hard-filter outcomes, component scores, evidence, and Safe-Call constraints.No findings

    Why it is in this group

    • Nothing in its name or description says what it does, so it is counted as a write.
    • The server marks it read-only (readOnlyHint: true), but an annotation never lowers a class.
    Description, as the server wrote it
    Recommend a provider with hard-filter outcomes, component scores, evidence, and Safe-Call constraints.

    Annotations the server declares

    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
    • readOnlyHint: true
Only reads 3Claude Code: allow
  • compare_providersCompare matching providers across capability, verification, safety, payment, freshness and unknowns.No findings

    Why it is in this group

    • The name "compare_providers" says "compare": a word for reading.
    Description, as the server wrote it
    Compare matching providers across capability, verification, safety, payment, freshness and unknowns.

    Annotations the server declares

    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
    • readOnlyHint: true
  • get_capabilityGet every provider capability with this canonical action ID, including evidence and Safe-Call contract.No findings

    Why it is in this group

    • The name "get_capability" says "get": a word for reading.
    Description, as the server wrote it
    Get every provider capability with this canonical action ID, including evidence and Safe-Call contract.

    Annotations the server declares

    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
    • readOnlyHint: true
  • get_provider_profileGet a canonical provider profile with capabilities, evidence references, and safe-call contracts.No findings

    Why it is in this group

    • The name "get_provider_profile" says "get": a word for reading.
    Description, as the server wrote it
    Get a canonical provider profile with capabilities, evidence references, and safe-call contracts.

    Annotations the server declares

    • destructiveHint: false
    • idempotentHint: true
    • openWorldHint: false
    • readOnlyHint: true

This is not a safety rating, and no finding is not a safety guarantee. The groups and findings come from fixed checks of the tool definitions the server returned on the date shown, not from its code or what it does when called. How the checks work.

Install

Pinned to version 0.1.0. Values in angle brackets are yours to fill in.

claude mcp add --transport http trust-live https://mcp-selection-lab-production.up.railway.app/trust-live/mcp

Run in a terminal.

Claude Code permissions

Allow the reads, ask before a change, deny the deletes. Paste into .claude/settings.json.

.claude/settings.json
{
  "permissions": {
    "allow": [
      "mcp__trust-live__compare_providers",
      "mcp__trust-live__get_provider_profile",
      "mcp__trust-live__get_capability"
    ],
    "ask": [
      "mcp__trust-live__recommend_provider",
      "mcp__trust-live__evaluate_safe_call"
    ],
    "deny": []
  }
}

Tools that only read are allowed, tools that change things ask first, tools that delete or cancel are denied. It assumes the server is installed as trust-live. Review it before you use it.