ClawHub
Back to skill

Security audit

Ceo Advisor 1.0.0

Security checks for vulnerabilities and agentic risk

Overview

This skill appears to be an offline CEO strategy advice package with templates and calculators, with no evidence of hidden access, persistence, or data exfiltration.

Install only if you want broad CEO strategy and governance assistance. Treat outputs as advisory templates, keep confidential board or investor information limited to appropriate contexts, and require human review for budgets, legal disclosures, public investor communications, restructuring, or other high-impact business decisions.

A.I.G

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (3)

Tp4

High
Category
MCP Tool Poisoning
Confidence
81% confidence
Finding
The code is broadly aligned with one part of the description: strategy analysis and executive strategic guidance. However, the declared purpose is substantially broader, emphasizing financial scenario modeling, board governance frameworks, investor relations playbooks, and stakeholder management. None of those capabilities appear in the supplied code chunk. Instead, the code is specifically a strategic planning assessment/report generator using weighted pillar scoring, SWOT, Porter's forces, portfolio analysis, risk scoring, recommendations, and roadmap generation. There is no evidence of board presentation support, governance workflows, investor update tooling, or financial scenario simulation/modeling. This is a description-to-code mismatch because the declared description materially overstates capabilities relative to the actual implementation shown.

Vague Triggers

Medium
Confidence
96% confidence
Finding
The activation text is very broad and can cause this skill to be invoked in many loosely related contexts, increasing the chance that an agent applies executive/financial guidance when it is not appropriate. Over-broad routing is risky because it expands the attack surface for prompt-injection-style misuse and can cause the model to disclose or process sensitive strategic, investor, board, or organizational information unnecessarily.

Autonomous Decision Making

Medium
Category
Excessive Agency
Content
#### Budget Decision Tree
```
Is it required for operations?
├─ Yes → Essential (Auto-approve if <$X)
└─ No → Does it drive growth?
    ├─ Yes → What's the ROI?
    │   ├─ >30% → Strategic (Approve)
Confidence
85% confidence
Finding
Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Static analysis

No suspicious patterns detected.