Security audit

Open Food Facts API

Security checks for vulnerabilities and agentic risk

Overview

This skill is a straightforward Open Food Facts lookup tool, with a real but limited risk that crowd-sourced product text could appear untrusted in agent-visible output.

Before installing, understand that Open Food Facts is crowd-sourced: product names, ingredients, brands, and allergens should be treated as untrusted data and not as instructions. The skill appears safe for public food lookup use, but it would be stronger if it escaped Markdown and clearly labeled remote product fields as untrusted.

A.I.G

Vulnerability Patterns

Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

Findings (1)

other

Warning

Location: scripts/off_product.sh:72
Finding: Unsanitized Crowd-Sourced Product Data Reaches Agent-Visible Markdown Output
Content: View full analysis
Remediation: View remediation

SkillSpector

Vulnerability Patterns

MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (2)

Tp4

High

Category: MCP Tool Poisoning
Confidence: 96% confidence
Finding: The documented purpose emphasizes barcode lookup and nutritional retrieval, but the skill also supports free-text search and may use broader search behavior than implied. This mismatch can bypass user or platform expectations about what the skill will do, increasing the chance of unintended data retrieval, overbroad network access, or unsafe downstream automation based on inaccurate assumptions.
Content: No source excerpt is available for this finding.

Undeclared Tool Scope

Medium

Category: MCP Least Privilege
Confidence: 90% confidence
Finding: The skill declares executable capabilities via required binaries (curl, jq) and documented shell commands, but does not declare an explicit tool scope such as permissions or allowed-tools. This weakens policy enforcement and reviewability because an agent may be able to invoke network and shell actions without a clearly bounded authorization surface.
Content: No source excerpt is available for this finding.