Ceo Delegation

• Outcome: Review
• Latest audit: May 28, 2026 at 9:27 AM UTC
• Version:
  1.0.0

Overview

This skill is not malicious, but it broadly routes tasks through sub-agents and memory without clear limits or consent controls.

Install only if you intentionally want an assistant to delegate most work to sub-agents. Set explicit limits for sub-agent creation, cost, time, retries, installations, file changes, and memory writes, and avoid using it for secrets or sensitive work unless you are comfortable with that content being shared with sub-agents or stored.

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: 88% confidence
• Finding: The workflow instructs use of memory_search/RAG to retrieve prior cases even though the stated purpose is simple delegation and monitoring. This expands data access beyond what users would reasonably expect, creating unnecessary exposure of prior task data, secrets, or unrelated context to delegated workflows.

Context-Inappropriate Capability

Medium

• Confidence: 91% confidence
• Finding: Automatically recording outcomes into memory introduces persistent storage of potentially sensitive task content without clear need for a delegation-only skill. This can lead to privacy leakage, retention of secrets, and unintended reuse of sensitive outputs in future tasks.

Context-Inappropriate Capability

Medium

• Confidence: 87% confidence
• Finding: The documentation broadens the skill into installation, configuration, and file-operation domains, materially expanding operational scope beyond delegation. That mismatch can cause the agent to invoke more powerful actions than users expect, increasing the chance of unsafe changes to systems or data through delegated subagents.

Vague Triggers

High

• Confidence: 95% confidence
• Finding: The trigger condition applies the workflow to 'any task,' making the skill activate far too broadly. In context, that means arbitrary user requests may be automatically delegated to subagents, potentially causing unnecessary data sharing, privilege expansion, or unsafe execution patterns even when delegation is inappropriate.

VirusTotal

66/66 vendors flagged this skill as clean.