RSAD Copilot

Security checks across malware telemetry and agentic risk

Overview

This is a documentation-only helper for IBM RSAD/FTM design work, with one overly broad activation phrase but no hidden code or data access.

Install this only if you work on IBM RSAD/RSA and FTM design tasks. Review generated DB2 SQL and design changes through normal enterprise controls, and do not rely on the broad trigger wording for unrelated payment, database, or operational work.

SkillSpector

By NVIDIA

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

High

Confidence: 93% confidence
Finding: The trigger logic is explicitly overbroad and includes 'When in doubt, trigger this skill,' which encourages invocation even when user intent is ambiguous. That can cause the agent to route unrelated prompts into this skill, increasing the chance of incorrect actions, disclosure of design-specific guidance in the wrong context, and prompt-surface expansion for downstream misuse.

VirusTotal

60/60 vendors flagged this skill as clean.

View on VirusTotal