Description-Behavior Mismatch
Medium
- Category
- Not specified by scanner
- Confidence
- 93% confidence
- Finding
L003将技能描述为定义规则、搜索策略、信息源、提纯标准、自检清单和输出模板，语义上偏向方法论/规范文档。但正文在L042、L155明确要求访问本地桌面目录、在不存在时创建目录，并将结果写入本地文件，这属于具体的数据读写执行行为而不仅是规范说明。
- Content
Security audit
Daily AI Briefing（每日AI简报）
Security checks for vulnerabilities and agentic risk
Overview
This skill is a Chinese daily AI briefing workflow that searches public AI sources and saves a local markdown briefing, with narrow and disclosed local file use.
Install only if you are comfortable with a Chinese-language workflow that searches the web and keeps briefing history in ~/Desktop/daily_ai_briefing. Review that folder if prior briefings might contain sensitive notes, and change the path or ask the agent not to save files if you do not want local retention.
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
- 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
- Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (4)
Natural-Language Policy Violations
Medium
- Category
- Not specified by scanner
- Confidence
- 96% confidence
- Finding
The instruction '全文必须使用专业、流畅的中文输出' forces a specific language for all outputs. Under the policy, language constraints are only acceptable when the skill offers user choice or clearly documents a justified region-specific need, which is not present here.
- Content
Missing User Warnings
Medium
- Category
- Not specified by scanner
- Confidence
- 95% confidence
- Finding
The skill instructs the agent to read from and create files under a fixed desktop directory without explicit user notice or consent. This creates a privacy and transparency risk because local files may contain historical content or sensitive information, and the user is not clearly warned that persistent local state will be accessed and created.
- Content
Missing User Warnings
Medium
- Category
- Not specified by scanner
- Confidence
- 97% confidence
- Finding
Writing the generated briefing to a predetermined local path without an explicit warning or opt-in introduces unauthorized persistence. Even if the content is benign, silent writes can surprise users, overwrite expectations about local state, and leave sensitive research traces on disk.
- Content
Static analysis
No suspicious patterns detected.