Back to skill

Security audit

og-previews

Security checks for vulnerabilities and agentic risk

Overview

This skill is a straightforward link-preview helper that checks public page metadata through getwowlink and guides small, relevant code edits.

Install only if you are comfortable using getwowlink for link-preview checks. Use it on public pages or staging pages safe to share, and review any generated meta-tag edits before deployment.

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

The listed trigger phrases include generic everyday requests such as "check my previews" and "add OG images" without defining clear boundaries for when this skill should or should not activate. In a markdown skill description, this can make invocation overly broad because the phrases are not constrained by explicit exclusions or a narrow activation context.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
84% confidence
Finding

The hardcoded external endpoint normalizes third-party transmission as part of the workflow, with no safeguards around sensitive targets or confidential parameters embedded in URLs. In context, the skill is meant for website preview checking, which makes the behavior understandable, but still risky if applied to non-public pages or internal environments.

Content

Scanner excerpt · SKILL.md (reported line 28)May include surrounding context.

Without it, use the public API:

bash
curl -s -X POST https://api.getwowlink.com/tools/og-check -H 'content-type: application/json' -d '{"url":"example.com/page"}'

or open https://www.getwowlink.com/og-checker?url=<page> for the visual report.

External Transmission

Medium
Category
Data Exfiltration
Confidence
84% confidence
Finding

The hardcoded external endpoint normalizes third-party transmission as part of the workflow, with no safeguards around sensitive targets or confidential parameters embedded in URLs. In context, the skill is meant for website preview checking, which makes the behavior understandable, but still risky if applied to non-public pages or internal environments.

Content

Scanner excerpt · SKILL.md (reported line 28)May include surrounding context.

Without it, use the public API:

bash
curl -s -X POST https://api.getwowlink.com/tools/og-check -H 'content-type: application/json' -d '{"url":"example.com/page"}'

or open https://www.getwowlink.com/og-checker?url=<page> for the visual report.

Static analysis

No suspicious patterns detected.