Skip to content

TrustScan

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

Security scanner for MCP servers and skills: Unicode injection, patterns, secrets.

Changes · 2
CompareRegistry entry

4 tools: 2 change things and 2 only read.

Scanned Oct 21, 2026

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

claude mcp add --transport http trust-scan https://trust-scan-production.up.railway.app/mcp/

Tools

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

4 tools

Changes things 2Claude Code: ask
  • trust_scan_fileSecurity-scan a single file for invisible Unicode, dangerous patterns, and secrets.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
    Security-scan a single file for invisible Unicode, dangerous patterns, and secrets. Returns a severity-weighted score and per-finding detail (rule, severity, location). Read-only: the file is never modified.

    Annotations the server declares

    • destructiveHint: false
    • idempotentHint: true
    • readOnlyHint: true
    • title: Scan Single File
  • trust_scan_serverSecurity-scan an MCP server or skill package before trusting it.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
    Security-scan an MCP server or skill package before trusting it. Runs all four checks — invisible Unicode prompt-injection, dangerous code patterns (MCP001–006), hardcoded secrets, typosquat package names — and returns a 0-100 score, letter grade, and detailed findings. Run this on any directory BEFORE wiring it into your agent. Read-only: never modifies the scanned target.

    Annotations the server declares

    • destructiveHint: false
    • idempotentHint: true
    • readOnlyHint: true
    • title: Scan MCP Server or Skill Package
Only reads 2Claude Code: allow
  • read_skillRead a product skill file by its skill:// URI.No findings

    Why it is in this group

    • The name "read_skill" says "read": a word for reading.
    Description, as the server wrote it
    Read a product skill file by its skill:// URI.
  • skills_list_toolList this product's skills.No findings

    Why it is in this group

    • The name "skills_list_tool" says "list": a word for reading.
    Description, as the server wrote it
    List this product's skills. Each entry carries the SKILL.md URI, its name and description, verbatim frontmatter, and a per-file sha256 manifest. Read a body with `read_skill`.

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-scan https://trust-scan-production.up.railway.app/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-scan__skills_list_tool",
      "mcp__trust-scan__read_skill"
    ],
    "ask": [
      "mcp__trust-scan__trust_scan_server",
      "mcp__trust-scan__trust_scan_file"
    ],
    "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-scan. Review it before you use it.