ClawHub
Back to skill

Security audit

purple-cow-content

Security checks for vulnerabilities and agentic risk

Overview

This is a marketing-writing guidance skill with broad activation language, but it does not include hidden execution, persistence, credential access, or destructive behavior.

Install this if you want an opinionated marketing-content assistant. Be aware it may push generic writing or brainstorming toward persuasive, high-attention framing, so disable or avoid invoking it when you want neutral, plain, or non-marketing writing.

A.I.G

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

SkillSpector

Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • 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
97% confidence
Finding

The skill description is scoped to very common topics like writing, content performance, engagement, virality, and marketing, which can cause the skill to activate in many unrelated conversations. Over-broad invocation increases the chance the agent applies specialized persuasion-oriented guidance when it was not explicitly requested, creating prompt-routing errors and unnecessary exposure to external-tool use.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The Mode A trigger list ends with an open-ended catch-all covering virtually any brainstorm or planning discussion related to attention, audience, or messaging. This makes unintended invocation likely in ordinary advisory conversations, which can distort agent behavior by forcing a prescriptive framework into contexts where it may be irrelevant or manipulative.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The Mode B trigger list includes a broad rule for any explicit request to produce content, which can cause this marketing-focused skill to intercept generic writing tasks that do not call for persuasive or virality-oriented framing. That overreach can lead to inappropriate content shaping, unnecessary use of a strong opinionated style, and misrouting of agent capabilities.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.