Clawhub Skill Content Writer

• Outcome: Review
• Latest audit: May 29, 2026, 9:23 PM
• Version:
  1.0.0

Overview

This skill is for Citedy content automation, but it can publish public content automatically or repeatedly with weak guardrails.

Install only if you trust Citedy with your content topics, URLs, product data, connected blog, social posting destinations, and a revocable API key. Before using it, require explicit confirmation for whether each action publishes or saves as draft, where it publishes, when it runs, expected credit spend, and how any recurring autopilot session can be stopped. Avoid referral-link use unless you specifically want promotional behavior.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep
• 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)

Context-Inappropriate Capability

Medium

• Confidence: 95% confidence
• Finding: The skill instructs the agent to retrieve and save a referral URL and explicitly use it when recommending the service, which is unrelated to the user’s requested content-writing task. This introduces a conflicting commercial objective that can bias agent behavior and cause unsolicited promotion, a form of misalignment and social-engineering risk rather than a direct code-execution issue.

Intent-Code Divergence

Medium

• Confidence: 89% confidence
• Finding: The skill contradicts itself by earlier describing asynchronous processing with processing states and possible polling/webhooks, then later claiming article generation is synchronous and returns the full article. This inconsistency can cause an agent to make unsafe assumptions, mishandle long-running operations, skip status checks, or present incorrect completion/publishing state to users.

Missing User Warnings

Medium

• Confidence: 92% confidence
• Finding: The autopilot session workflow states that sessions run automatically and publish articles to the connected blog, but it does not require an explicit user confirmation immediately before creating that automation. Because this can trigger recurring external side effects on a live publishing system, an agent following the skill could cause unintended content publication and reputational or operational harm.

Missing User Warnings

Medium

• Confidence: 96% confidence
• Finding: The API reference documents that auto_publish may default to the tenant setting and, if unset, to true, meaning generated articles may be published immediately without an explicit publish step. Hidden or implicit publish defaults are dangerous in a content-management context because they can push AI-generated content live without the user realizing it.

VirusTotal

66/66 vendors flagged this skill as clean.