ClawHub
Back to skill

Security audit

clearml-metrics-logging-pattern

Security checks for vulnerabilities and agentic risk

Overview

This skill is a ClearML metrics-logging guide with expected package-install and telemetry behavior for experiment tracking, though users should understand what ClearML will record.

Install only if you intend to use ClearML/expflow for experiment tracking. Review your ClearML configuration first, avoid logging sensitive dataset details or secrets as metrics, and prefer a self-hosted or appropriately governed ClearML server for private experiments.

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
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (1)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
87% confidence
Finding

The skill explicitly instructs users to install and use ClearML-based metrics logging, including reporting training, validation, and system telemetry such as GPU memory and learning rate, but it does not clearly warn that these values are sent to an external experiment-tracking service. In an MLOps context this can lead to unintentional disclosure of operational metadata, experiment details, and potentially sensitive run information if users assume logging is local-only.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.