Back to skill

Security audit

Gate Exchange Affiliate Program Skill

Security checks for vulnerabilities and agentic risk

Overview

This is a read-only Gate affiliate reporting skill whose credentials and API access match its stated purpose, with a minor trigger-word ambiguity to watch.

Install this only if you intend to use Gate affiliate or partner reporting through a configured Gate MCP session. Use a least-privilege Gate API key limited to Rebate:Read, and be careful with vague prompts containing "commission" because the agent should confirm Gate affiliate intent before querying account-specific data.

Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (1)

Vague Triggers

Medium
Confidence
92% confidence
Finding
The trigger phrase "commission" is so broad that it can match many unrelated finance or compensation requests, causing this credentialed affiliate skill to activate outside its intended scope. In context, that matters because the skill has access to authenticated rebate data and can query partner-only APIs, so accidental routing may expose sensitive affiliate information or cause inappropriate tool use.

Static analysis

No suspicious patterns detected.