Security audit
Amazon Research by Gecho
Security checks for vulnerabilities and agentic risk
Overview
This skill is a clearly scoped Amazon research connector that uses Gecho's MCP tools and logged-in browser setup to read product and review data, with no artifact evidence of hidden mutation or exfiltration behavior.
Before installing, understand that this skill requires trusting Gecho's Chrome extension and MCP server with access to the Amazon pages available in your logged-in browser session. Use it for read-only product and review research, avoid running it in sensitive Chrome profiles if unnecessary, and choose save directories deliberately because raw product or review results may be written locally.
Vulnerability Patterns
- 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
- Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Static analysis
No suspicious patterns detected.