ClawHub
Back to skill

Security audit

Ai Csuite

Security checks for vulnerabilities and agentic risk

Overview

This skill is a local business decision helper that reads company context and writes decision markdown, with no evidence of hidden networking, credential access, or system persistence.

Install only if you want a local script-backed strategy workflow. Use redacted or non-confidential company data when possible, review generated files before sharing or committing them, and be aware that logs/latest-decision.md may overwrite or retain sensitive decision context.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • 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
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (5)

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding
This code chunk does not implement the declared primary skill behavior. Instead of orchestrating an executive debate or generating business strategy outputs, it performs static scanning of files for suspicious patterns and non-stdlib imports. While the description briefly mentions security scripts for safe distribution, the provided code is only that auxiliary security scanner and not the main debate system described. Because the actual code’s primary purpose is materially different from the declared skill purpose, this is a description-behavior mismatch.

Ae1

High
Category
analysis-evasion
Content
- OpenSkills-compatible runners that support `SKILL.md`
Confidence
100% confidence
Finding
Referenced artifact was not completely inspected

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The README explicitly tells users to populate a local company configuration with real business data and generate decision artifacts, but it does not warn that those outputs and logs may contain sensitive strategic, financial, and operational information. In a skill designed for executive planning, generated markdown and local logs are especially likely to include confidential company context that could later be committed, shared, or exposed unintentionally.

Vague Triggers

Medium
Confidence
90% confidence
Finding
The activation text is broad enough to trigger on common strategic-advice requests, which can cause the skill to run local scripts in situations where the user only wanted conversational guidance. In this skill context, overbroad activation is more dangerous because the workflow includes Bash execution and file writes, increasing the chance of unintended code execution and side effects.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The instructions direct the agent to write `logs/latest-decision.md` and build session artifacts without telling the user beforehand that filesystem changes will occur. This is dangerous because users may unknowingly permit persistent local modifications, and in tool-enabled environments that can create unwanted files, overwrite prior outputs, or leak sensitive business context into local logs.

Static analysis

No suspicious patterns detected.