ClawHub
Back to skill

Security audit

SlideClaw

Security checks for vulnerabilities and agentic risk

Overview

This skill is a coherent SlideClaw/Marp helper; its main risk is the disclosed npm/npx CLI execution path, not hidden behavior.

Install this if you trust the SlideClaw npm package and are comfortable with an agent using it to create and render slide files. For stricter environments, install a reviewed or pinned SlideClaw version first so the skill uses the local binary instead of npx latest, and disable copying to Downloads with SLIDECLAW_NO_DOWNLOADS=1 or --no-copy-to-downloads if you want outputs kept only in the workspace.

SkillSpector

Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (1)

Context-Inappropriate Capability

Medium
Confidence
96% confidence
Finding
The skill explicitly instructs the agent to fall back to `npx -y slideclaw@latest`, which downloads and executes the latest package version from a remote registry at runtime. That creates a supply-chain and arbitrary code execution risk in the user's environment, especially because the command is part of the required bootstrap flow and may run without meaningful user review.

Static analysis

No suspicious patterns detected.