Back to skill

Security audit

Fetch

Security checks across malware telemetry and agentic risk

Overview

This skill is a straightforward public URL fetching helper that saves fetched pages locally, with no evidence of credential access, exfiltration, persistence beyond local history, or deceptive behavior.

Install this only if you want an agent helper that can fetch web pages and save copies locally. Use it with deliberate public http/https URLs, avoid internal or credential-bearing URLs, and periodically review or clear the saved history under ~/.openclaw/workspace/memory/fetch if the fetched content may be sensitive.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (2)

Lp3

Medium
Category
MCP Least Privilege
Confidence
92% confidence
Finding
The skill advertises and documents network access plus local file read/write behavior, but the metadata does not declare permissions explicitly. That mismatch weakens review and policy enforcement because a caller or platform may underestimate what the skill can do, especially since it stores fetched content and job history on disk. The stated purpose is benign, but undeclared capabilities increase the risk of misuse and accidental overreach.

Vague Triggers

Medium
Confidence
84% confidence
Finding
The invocation description is very broad: 'fetch, download, inspect, clean, or save content from a public URL' can match many ordinary user requests and trigger the skill in situations where users did not explicitly ask for network retrieval or local persistence. Because the skill performs both remote fetching and local storage, over-broad routing can cause unintended data access, persistence, or exposure to untrusted remote content more often than necessary. In context this is not overtly malicious, but it increases the attack surface.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.