ClawHub
Back to skill

Security audit

LYGO API Token Saver

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed token-saving workflow guide, but users should review its broad activation phrases and API-key routing before relying on it.

Install only if you want an agent to prefer local tools, compact responses, and alternate/model-specific API routing to reduce token spend. Confirm that the referenced vault, companion skills, and local scripts are yours and trusted, and use explicit wording when you want paid frontier models or push-related actions.

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 (2)

Vague Triggers

Medium
Confidence
91% confidence
Finding
The description embeds multiple broad natural-language triggers such as 'token saver', 'save tokens', 'API budget', and 'before long ... calls', which can cause the skill to activate in unrelated conversations where the user is only discussing cost or planning model usage. Because this skill changes model/key-routing behavior and encourages alternative execution paths, unintended activation can alter security-sensitive runtime behavior without clear user consent.

Vague Triggers

Medium
Confidence
94% confidence
Finding
The user phrase mapping uses ambiguous triggers like 'token saver', 'pay to go', and especially 'push all' to switch into behavior modes. These phrases are common and context-dependent, so the skill may activate or change execution strategy during ordinary conversation, potentially causing unintended use of alternate API keys, local tools, or reduced-review workflows.

VirusTotal

62/62 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.