Multi Platform Publisher Claw

• Outcome: Pass
• Latest audit: May 29, 2026 at 8:41 PM UTC
• Version:
  1.0.0

Overview

This skill is a disclosed 96Push publishing helper with strong confirmation gates for public posting and account-changing actions.

Install only if you use and trust 96Push and understand that approved actions may publish publicly or modify logged-in social media accounts. Keep PUSH_API_KEY in a local secret/env file, review the exact accounts and content before approving, and prefer draft or single-account tests first.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• MCP Least Privilege: Underdeclared Capability, Wildcard Permission, Missing Permission Declaration
• 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: 95% confidence
• Finding: The skill clearly instructs execution of a publishing script and describes networked multi-platform posting plus use of credentialed platform accounts, yet no permissions are declared. This creates a dangerous trust gap: the skill can perform external actions and likely write files or logs without transparent consent boundaries, increasing the chance of unintended posting or misuse of credentials.

Vague Triggers

Medium

• Confidence: 87% confidence
• Finding: The trigger phrases are broad terms like '发布内容', 'SEO优化', and '自动发布', which can match common user requests and cause the skill to activate in situations where the user did not intend cross-platform automated posting. In a skill that can publish externally, overbroad activation materially raises the risk of accidental execution and unintended dissemination.

Missing User Warnings

Medium

• Confidence: 92% confidence
• Finding: The workflow proceeds directly to execution of automated cross-platform publishing without an explicit warning or confirmation about irreversible external actions. Because this skill can post content to multiple public platforms, missing a prominent user warning increases the chance of accidental publication, reputational damage, and distribution of sensitive or incorrect content.

Missing User Warnings

Low

• Confidence: 90% confidence
• Finding: Referencing `.env`-based platform account configuration without warning about secret handling encourages casual placement of sensitive credentials in local files without guidance on protection. In a skill that automates posting to external services, exposed tokens or passwords could let an attacker hijack publishing accounts or exfiltrate private account access.

VirusTotal

66/66 vendors flagged this skill as clean.