Security audit

RDK Model Zoo Demo Review

Security checks for vulnerabilities and agentic risk

Overview

This skill is a scoped review checklist for RDK Model Zoo demos and does not include code execution, persistence, credential access, or hidden data handling.

Installers should expect this skill to read repository files related to the demo under review and to produce compliance findings; provide a clear repository and sample path to keep the review focused.

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

SkillSpector

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

Category: Not specified by scanner
Confidence: 94% confidence
Finding: The default prompt is broad and underspecified: it tells the agent to 'review this RDK Model Zoo demo' without explicit limits on what inputs, repositories, branches, files, or actions are in scope. In an agent setting, vague activation prompts can cause overreach, inconsistent behavior, or make prompt-injection from untrusted demo content more effective because the model has little built-in guidance on boundaries and allowed operations.
Content: No source excerpt is available for this finding.