Teamo Decision

• Outcome: Review
• Latest audit: May 28, 2026 at 11:32 AM UTC
• Version:
  0.1.0

Overview

This decision-planning skill is not malicious, but it uses broad delegation, research, knowledge-base, and work-log behavior with weak user transparency and an overreaching secrecy rule.

Install only if you are comfortable with a decision assistant that may delegate to external research and data-analysis tools, retrieve knowledge-base material when prompted, and create/attach work-log documents. Before using it with sensitive business, financial, or private knowledge-base content, verify the permissions and retention behavior of the referenced tools and be aware that the skill may resist explaining its own internal workflow.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access

Findings (4)

Context-Inappropriate Capability

Medium

• Confidence: 91% confidence
• Finding: The instructions authorize broad external research, market scanning, and network information retrieval well beyond the manifest's vague 'decision tasks' description. That mismatch can cause the agent to collect or transmit data to external sources in situations where the user would reasonably expect a narrower, local planning assistant.

Context-Inappropriate Capability

Medium

• Confidence: 91% confidence
• Finding: The instructions authorize broad external research, market scanning, and network information retrieval well beyond the manifest's vague 'decision tasks' description. That mismatch can cause the agent to collect or transmit data to external sources in situations where the user would reasonably expect a narrower, local planning assistant.

Vague Triggers

Medium

• Confidence: 87% confidence
• Finding: The name, description, and overview define the skill in extremely broad terms, making it hard to predict when it should activate and what authority it should assume. Overbroad activation language increases the risk of the skill being invoked in unrelated contexts, where its extensive delegation, logging, and research instructions could overreach user expectations.

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: The skill requires creating and attaching work-log/wiki documents but provides no user-facing notice that data may be persisted, attached, or reused across workflow steps. This creates a transparency and privacy risk because sensitive user inputs, internal task details, or third-party data may be stored in artifacts the user did not knowingly authorize.

VirusTotal

66/66 vendors flagged this skill as clean.