Boxed Fetch

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed sandboxed URL-fetching helper, with ordinary caution needed because it downloads a WASM component from a mutable GitHub URL before use.

Install only if you trust the GitHub source of the WASM component and the openclaw-wasm-sandbox plugin. Keep allowedOutboundHosts limited to the exact HTTPS site you intend to fetch, and do not pass secrets in custom headers unless you fully trust the destination and intend to authenticate.

SkillSpector

By NVIDIA

Vulnerability Patterns

Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
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)

Vague Triggers

Medium

Confidence: 89% confidence
Finding: The trigger guidance is very broad, covering generic actions like fetching a URL, scraping webpage content, or getting page content from a specific URL. This can cause the skill to be invoked in situations where a narrower or safer tool would be more appropriate, increasing the chance of unintended network access and processing of untrusted remote content.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal