T08 · Insecure Dependencies
Warning
- Location
SKILL.md:40- Finding
Unpinned Third-Party Python Dependencies
- Content
View full analysis
- Remediation
View remediation
Security audit
AI销量预测助手
Security checks for vulnerabilities and agentic risk
Overview
This sales forecasting skill is purpose-aligned, with no evidence of malware, but users should review its unpinned Python installs, external model download, and missing packaged scripts before relying on it.
Install only in an isolated virtual environment, consider pinning dependency versions, and be aware that first use downloads a large model from an external endpoint. Do not run this on sensitive business data until you confirm the actual forecasting scripts and data-handling behavior, since the package reviewed here only includes SKILL.md and not the referenced scripts.
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 (1)
Vulnerability Patterns
- Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
- Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
- Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
- 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
- 90% confidence
- Finding
The description says to use the skill when the user asks about '销量预测、销售预测、趋势预测、时序预测', which includes broad phrases like '趋势预测' and '时序预测' that could overlap with many general forecasting or trend-analysis requests. It does not provide explicit constraints, exclusions, or negative examples clarifying when the skill should not activate.
- Content
Natural-Language Policy Violations
Medium
- Category
- Not specified by scanner
- Confidence
- 95% confidence
- Finding
The manifest description and document content are entirely in Chinese, presenting the skill as a Chinese-language assistant without indicating that users may choose another language. This can violate language/locale policy when a skill implicitly forces a specific language without opt-in.
- Content
Missing User Warnings
Low
- Category
- Not specified by scanner
- Confidence
- 79% confidence
- Finding
The file documents use of a HuggingFace mirror endpoint and notes that the model will be downloaded on first run, but it does not clearly warn users about the network activity or advise them to consider data-handling/privacy implications in restricted environments. Because the skill processes user-supplied sales data and references external network endpoints, an explicit user-facing warning would improve safety disclosure.
- Content
Static analysis
No suspicious patterns detected.