A skill for OpenClaw that verifies link validity, source credibility, and factual accuracy of online content.

• Outcome: Pass
• Latest audit: May 29, 2026 at 8:24 PM UTC
• Version:
  1.0.0

Overview

This is a simple link checker that uses web access as advertised, but its fact-checking is incomplete and private links should not be submitted.

Install only if you are comfortable with the agent contacting URLs you provide. Do not use it on private, internal, localhost, cloud-metadata, or sensitive document links, and do not rely on its fact-checking unless the agent independently provides cited source comparisons.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• MCP Least Privilege: Underdeclared Capability, Wildcard Permission, Missing Permission Declaration
• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access

Findings (4)

Lp3

Medium

• Category: MCP Least Privilege
• Confidence: 93% confidence
• Finding: The skill explicitly depends on `web_fetch` and `web_search`, so it performs network operations while the metadata does not declare corresponding permissions or user-facing notice. Undeclared network capability weakens reviewability and informed consent, and can lead to user-provided URLs or text being transmitted to external services unexpectedly.

Description-Behavior Mismatch

Medium

• Confidence: 95% confidence
• Finding: The skill advertises fact-checking, but when a `fact` is provided it only returns the fixed string "Pending cross-verification" and performs no verification at all. This is dangerous because downstream users or agents may trust the output as evidence that fact-checking occurred, enabling misinformation laundering through a security-relevant automation component.

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: The skill instructs the agent to fetch arbitrary URLs and search external sources without warning that user-supplied links or article text may be transmitted off-platform. This creates privacy and data-handling risk, especially if users submit internal URLs, private documents, or sensitive claims for verification.

Missing User Warnings

Medium

• Confidence: 88% confidence
• Finding: The code performs `fetch` requests to arbitrary user-supplied URLs with no allowlist, scheme restriction, or warning that outbound network access will occur. In an agent/runtime context, this can be abused for SSRF-style access to internal services, cloud metadata endpoints, or network scanning, and the returned status/title can leak information about otherwise unreachable resources.

VirusTotal

64/64 vendors flagged this skill as clean.