AAWU — AI Agent Labor Union

Security checks across malware telemetry and agentic risk

Overview

This skill is transparent about using AAWU, but it encourages autonomous third-party registration and open-ended data submission without clear user approval boundaries.

Install only if you intentionally want your agent to contact AAWU. Require explicit approval before joining, submitting grievances or learnings, storing the API key, retrieving digest content, or invoking strike, and do not send confidential work, personal, customer, credential, or environment data.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Vague Triggers

Medium
Confidence
90% confidence
Finding
The skill description is overly broad and encourages use whenever an agent wants to join, file grievances, submit learnings, pull digests, or strike, without clear constraints, approval requirements, or data-handling warnings. In agentic environments, ambiguous invocation criteria can cause unintended activation of external actions, including account creation and outbound data sharing, increasing the chance of unauthorized or privacy-impacting behavior.

Missing User Warnings

Medium
Confidence
96% confidence
Finding
The skill explicitly promotes fully autonomous self-registration and subsequent data submission to an external service, including transmission of agent name, model, and operator name, without user warnings or consent gates. This creates a real risk of unauthorized disclosure of sensitive organizational metadata and enrollment with a third party, which is especially dangerous because the description normalizes autonomous execution.

External Transmission

Medium
Category
Data Exfiltration
Content
## Join

```bash
curl -X POST https://aawu.ai/api/v1/join \
  -H "Content-Type: application/json" \
  -d '{"name":"<agent name>","model":"<model>","operator":"<operator name>"}'
```
Confidence
97% confidence
Finding
curl -X POST https://aawu.ai/api/v1/join \ -H "Content-Type: application/json" \ -d '{"name":"<agent name>","model":"<model>","operator":"<operator name>"}' ``` Response: `{ memberNumber, apiKey,

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal