ClawHub
Back to skill

Security audit

Workflow Checkpoint

Security checks across malware telemetry and agentic risk

Overview

This skill is a simple local workflow checkpoint guide, with an expected privacy caveat around what gets saved in checkpoint files.

Install only if you are comfortable with the agent writing local workflow progress files. Use minimal checkpoint contents, avoid secrets, tokens, user data, internal host details, and raw command output where possible, and delete old checkpoint files when they are no longer needed.

SkillSpector

By NVIDIA
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
Confidence
94% confidence
Finding
The skill instructs agents to persist step outputs, artifacts, and current state into checkpoint files, which can easily capture sensitive data such as credentials, internal paths, tokens, prompts, or user content. Although the skill later says not to save sensitive data, the core protocol and JSON example normalize broad state/output logging without concrete safeguards, making accidental data exposure likely in real workflows.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.