Letta

Security checks across malware telemetry and agentic risk

Overview

This is a small documentation-style Letta skill with broad activation wording but no executable code, hidden behavior, or high-impact authority.

Install this if you want OpenClaw to provide Letta-oriented guidance for memory-augmented agent work. Be aware it may activate on broad memory-agent discussions, and only run the referenced npm install command when you explicitly want the Letta CLI installed.

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

Medium

Confidence: 90% confidence
Finding: The activation guidance is broad enough to trigger on generic discussion of AI agents, memory, or persistent context rather than a narrowly defined user intent. In a skill-routing system, this can cause unintended invocation, increase prompt-surface exposure, and let the skill influence responses outside its intended scope.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal