ClawHub
Back to skill

Security audit

Cts Anomaly Check

Security checks across malware telemetry and agentic risk

Overview

The skill appears purpose-built for Huawei Cloud CTS auditing, but it under-discloses credential and billing API access that users should review before installing.

Review this skill before installing. It may read Huawei Cloud audit logs, account debt status, user/resource operation details, and create temporary Huawei Cloud access keys during execution. Install only if you are comfortable granting IAM temporary credential creation plus BSS billing balance access in addition to CTS trace access, and run it with explicit region, project, account, time range, and output-log controls.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (4)

Lp3

Medium
Category
MCP Least Privilege
Confidence
91% confidence
Finding
The skill documentation declares only a CTS API permission, but the skill clearly instructs users to run a local Python script and the analyzer detected network and shell capabilities. This mismatch is dangerous because operators may approve or invoke the skill without understanding that it can execute code and make external calls, weakening review, consent, and sandboxing assumptions.

Description-Behavior Mismatch

Medium
Confidence
93% confidence
Finding
The skill advertises CTS anomaly inspection but also performs an unrelated BSS billing/debt query, which expands data access beyond user expectations and least-privilege needs. In an agent-skill context, this kind of hidden scope expansion is dangerous because it can silently collect or reveal sensitive account-status information that the user did not ask for.

Context-Inappropriate Capability

Medium
Confidence
95% confidence
Finding
The script creates temporary IAM access keys and uses a custom signer to call APIs directly, granting broader credentialed capability than is necessary for reading CTS traces through the existing CLI wrapper. In a skill environment, generating and handling temporary AK/SK increases the blast radius if the skill is modified, logs leak, memory is exposed, or additional unauthorized API calls are added later.

Vague Triggers

Medium
Confidence
84% confidence
Finding
The trigger phrases are broad everyday requests such as checking logs or looking for recent abnormal operations, which can cause the skill to activate unintentionally in unrelated contexts. In a security-audit skill, accidental invocation is more sensitive because it may initiate log analysis, network/API access, or expose operational security data when the user did not explicitly intend that action.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.