Ae1
High
- Category
- analysis-evasion
- Content
node scripts/generate_content_pack.js --input product.json --format markdown
- Confidence
- 100% confidence
- Finding
- Referenced artifact was not completely inspected
Security audit
Commerce Content Generator
Security checks for vulnerabilities and agentic risk
Overview
This skill is a local commerce copywriting helper for Chinese-platform content and does not show hidden data access, persistence, or destructive behavior.
Install this if you want product marketing drafts for Chinese social commerce channels. Review generated claims before publishing, especially for health, finance, legal, children, food, supplements, or high-priced products, and provide real evidence so the skill does not rely on assumptions.
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
- Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
- Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
- MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
- Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
- Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (11)
Natural-Language Policy Violations
Medium
- Confidence
- 90% confidence
- Finding
- The code defaults content generation to "xiaohongshu" when no platform is provided. This imposes a specific platform/locale choice automatically rather than offering an explicit user selection or opt-in, which is a natural-language policy concern under locale/language constraints.
Natural-Language Policy Violations
Medium
- Confidence
- 95% confidence
- Finding
- The script generates all platform names, hashtags, templates, and user-facing copy in Chinese, such as the Chinese platform labels and the generated content bodies. There is no option for users to choose language/locale and no documented justification that this is a China-specific or region-limited skill, so it effectively forces a specific language.
Natural-Language Policy Violations
Medium
- Confidence
- 93% confidence
- Finding
- This test asserts that generated content must include the Chinese term "证据", which semantically enforces a specific output language/locale. The file does not indicate any user choice or justified region-specific constraint, so this appears to violate the language/locale policy requirement.
Natural-Language Policy Violations
Medium
- Confidence
- 95% confidence
- Finding
- This assertion requires the markdown output to contain the Chinese heading "小红书", indicating the skill is expected to generate Chinese-localized platform content by default. Because no user opt-in or documented locale limitation is present in this file, the natural-language behavior appears to force a specific locale.
Natural-Language Policy Violations
Low
- Confidence
- 82% confidence
- Finding
- The instruction to mark explicit assumptions with `待补充证据` imposes a Chinese-language label in user-facing output without indicating user choice or opt-in. This can violate language/locale policy when users expect output in another language.
Vague Triggers
Low
- Confidence
- 88% confidence
- Finding
- The manifest defines a default prompt of "Create a platform-specific content pack for this product," but provides no explicit trigger constraints, exclusions, or narrow invocation context. In a manifest file, this kind of generic content-generation phrasing can be too broad because it could match ordinary user requests without clearly delimiting when the skill should or should not activate.
Natural-Language Policy Violations
Low
- Confidence
- 76% confidence
- Finding
- The description explicitly targets Xiaohongshu and Douyin outputs, which suggests a China-specific language/locale context, but it does not say this is optional or user-selected. Because SQP-3 applies to natural-language locale constraints, this can be read as forcing a specific locale without clear opt-in.
Natural-Language Policy Violations
Low
- Confidence
- 83% confidence
- Finding
- The package description states that the skill outputs Xiaohongshu notes and Douyin scripts, which are platform- and language/locale-specific formats, but it does not mention user choice or opt-in. This creates a mild natural-language policy concern because the skill appears to impose a specific locale/content context by default rather than explicitly offering alternatives.
Natural-Language Policy Violations
Low
- Confidence
- 78% confidence
- Finding
- The document structures guidance entirely around Chinese platforms such as Xiaohongshu, Douyin, Zhihu, and Moments, which implicitly constrains locale and platform context. Because it does not state that this is optional, region-specific, or user-selected, it can be read as forcing a specific locale/platform policy without opt-in.
Natural-Language Policy Violations
Low
- Confidence
- 87% confidence
- Finding
- Line L08 instructs authors to label missing evidence as `待补充证据`, which forces Chinese output terminology. The file does not offer a language choice or explain that the checklist is intended only for a Chinese-language workflow, so this can violate language/locale policy expectations.
Static analysis
Detected: suspicious.dangerous_exec
Shell command execution detected (child_process).
Critical
- Code
- suspicious.dangerous_exec
- Location
- test/generate_content_pack.test.js:59