other
Warning
- Location
SKILL.md:123- Finding
Guidance for Evading Moderation of Regulated Product Claims
- Content
View full analysis
- Remediation
View remediation
Security audit
domestic-marketing-compliance
Security checks for vulnerabilities and agentic risk
Overview
This text-only marketing compliance skill has no code or persistence, but it mixes compliance guidance with risky moderation-evasion substitutions, viral/deceptive templates, a public skin-photo prompt, and an unaudited companion skill requirement.
Review before installing. The skill is not executing code, but its advice can lead an agent to create risky marketing copy, disguise regulated claims, solicit public skin photos, or rely on an unaudited companion skill. It should be tightened to compliance review and safe rewrites only, with unsupported claims removed rather than rephrased.
Vulnerability Patterns
- Insecure DependenciesIntroduces malicious components through unsafe dependency sources
- 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
Findings (5)
other
Warning
- Location
SKILL.md:201- Finding
Solicitation of Potentially Sensitive Skin Photographs
- Content
View full analysis
- Remediation
View remediation
other
Warning
- Location
SKILL.md:229- Finding
Template Promotes an Unsupported Extreme Weight-Loss Claim
- Content
View full analysis
- Remediation
View remediation
other
Note
- Location
SKILL.md:244- Finding
Template Encourages Manufactured Urgency and Unverified Download Distribution
- Content
View full analysis
- Remediation
View remediation
T08 · Insecure Dependencies
Warning
- Location
SKILL.md:12- Finding
Mandatory Loading of an Unbundled and Unpinned Skill Dependency
- Content
View full analysis
- Remediation
View remediation
Vulnerability Patterns
- Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
- Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
- MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
- Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
- Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (6)
Intent-Code Divergence
High
- Category
- Not specified by scanner
- Confidence
- 97% confidence
- Finding
The document claims to prioritize authenticity and avoid exaggeration, but many headline examples rely on sensationalism, unrealistic benefit claims, appearance transformation, or urgency tactics. In a compliance skill, these examples normalize borderline or noncompliant copy patterns and may cause downstream generation of deceptive marketing content, especially in beauty, health, parenting, and finance-adjacent contexts.
- Content
Vague Triggers
Medium
- Category
- Not specified by scanner
- Confidence
- 91% confidence
- Finding
The description says to load the skill before writing any domestic marketing/promotional copy, and the usage note says it must load whenever a user asks to write domestic marketing copy. This defines a wide activation scope but does not provide explicit exclusion conditions or negative examples, making invocation boundaries ambiguous for mixed-purpose writing requests.
- Content
Natural-Language Policy Violations
Medium
- Category
- Not specified by scanner
- Confidence
- 88% confidence
- Finding
The file is written as a mandatory guide for 'domestic' marketing copy and instructs that the skill must be loaded for such requests, but it does not offer any opt-in or language/locale selection. Under the stated policy, forcing a specific language/locale context without user choice can be a policy violation unless clearly justified and documented as region-specific.
- Content
Description-Behavior Mismatch
Medium
- Category
- Not specified by scanner
- Confidence
- 93% confidence
- Finding
The skill is presented as a compliance-review guide, but a large section expands into growth-oriented copywriting methodology for Xiaohongshu. That scope drift is dangerous because users or downstream agents may trust the entire skill as compliance-safe, even though part of it optimizes persuasion and virality rather than legal/platform safety.
- Content
Intent-Code Divergence
Medium
- Category
- Not specified by scanner
- Confidence
- 94% confidence
- Finding
The skill explicitly warns against risky interaction patterns on Xiaohongshu, yet later provides a template encouraging users to post photos in comments for free skin analysis. This creates inconsistent guidance and can still drive engagement manipulation or quasi-diagnostic behavior in a sensitive beauty/skin context, increasing moderation, compliance, and consumer-protection risk.
- Content
Context-Inappropriate Capability
Medium
- Category
- Not specified by scanner
- Confidence
- 91% confidence
- Finding
The skill contains recommendation and click-through optimization tactics aimed at increasing reach and engagement, which is not aligned with a narrowly defined compliance function. This mismatch can cause an agent to privilege performance tactics under the guise of safety review, leading to manipulative or policy-evasive outputs.
- Content
Static analysis
No suspicious patterns detected.