ClawHub
Back to skill

Security audit

Cms Project Governance

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed project-governance helper that manages scoped project records, with no evidence of hidden credential use, exfiltration, destructive automation, or system persistence.

Install this when you want an agent to maintain project-governance state and QA decisions. Expect it to create or update project-local Docs records when you authorize that work, and review any dry run before archive, reset, or rebaseline actions. If you only need ordinary coding with clear requirements, use the execution skill instead.

A.I.G

By Tencent
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

By NVIDIA
Vulnerability Patterns
  • 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
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Vague Triggers

Medium
Confidence
82% confidence
Finding
The skill advertises many broad natural-language triggers, including common project-management phrases such as scope creep, rebaseline, and should I continue. This can cause accidental invocation in unrelated conversations, which is risky because the skill is an authority-setting governance layer that may steer decisions, read project state, or initiate structured workflow when the user did not explicitly request it.

Missing User Warnings

Medium
Confidence
89% confidence
Finding
The skill explicitly instructs the agent to create, update, archive, and move repository files as part of normal governance flow, but it does not require a clear user-facing confirmation at the moment those write operations occur. In a governance skill that can reshape project state and authoritative records, this can lead to unintended modification of repository contents, especially during archival, rebaseline, or packet maintenance workflows.

Natural-Language Policy Violations

Medium
Confidence
88% confidence
Finding
The documented commands hard-code `--language zh-CN` with no indication that language selection is user-driven or configurable. This can override user expectations, cause incorrect localization behavior, and in governance workflows may lead to misunderstandings or acceptance errors if stakeholders are not operating in Chinese.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.