Ae1
High
- Category
- analysis-evasion
- Content
- `SKILL.md` - 技能说明文档（本文件）
- Confidence
- 100% confidence
- Finding
- Referenced artifact was not completely inspected
Security audit
X Twitter Collector
Security checks for vulnerabilities and agentic risk
Overview
This skill is a straightforward X/Twitter report collector that uses the browser to gather public account posts, screenshots, links, and engagement data into local reports.
Before installing, be aware that the skill may use your logged-in browser session to view X pages and may save local reports and full-page screenshots containing public tweet content, links, engagement counts, and visible page context. Use it only for accounts you are permitted to review, and clean the configured report/media directories if you do not want archived copies retained.
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
- 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
- Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
- Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (7)
Natural-Language Policy Violations
Medium
- Confidence
- 94% confidence
- Finding
- The README states '中英双语（默认）', and the config example also defaults reporting language to bilingual output. This imposes a language behavior by default rather than letting the user choose or explicitly opt in, which matches the language/locale policy concern.
Missing User Warnings
Medium
- Confidence
- 93% confidence
- Finding
- The skill explicitly collects tweet content, links, analytics, and full-page screenshots, and later suggests archiving/report generation, but it does not present a clear up-front warning that this data will be captured and stored as part of operation. This creates a privacy and transparency issue because users may invoke the skill without realizing it produces persistent copies of public content and screenshots, which may include more page context than expected.
Natural-Language Policy Violations
Medium
- Confidence
- 94% confidence
- Finding
- The report settings hard-code output as bilingual with specific locales zh-CN and en-US. This is a natural-language locale policy concern because the file mandates languages rather than offering a user-selectable option or documenting a justified region-specific requirement.
Missing User Warnings
Low
- Confidence
- 79% confidence
- Finding
- This markdown file says the skill includes full-page screenshots and can export reports to PDF/Excel, which affect local user data and storage. Although later notes mention screenshot file size and cleaning the media directory, the description does not clearly warn up front that running the skill will create and retain local files containing collected content.
Natural-Language Policy Violations
Low
- Confidence
- 95% confidence
- Finding
- The skill defaults to bilingual Chinese-English output and describes bilingual collection/reporting as part of its normal behavior without making it an explicit opt-in choice. While not a severe security flaw, this can increase unnecessary processing and disclosure of transformed content, especially when users only need one language, and it reduces user control over how collected data is handled.
Natural-Language Policy Violations
Low
- Confidence
- 84% confidence
- Finding
- The template consistently requires paired Chinese and English headings, fields, and content placeholders throughout the file, such as separate Chinese and English tweet content and topic fields. This imposes a specific language/locale format by default, but the template does not offer user opt-in or explain why a bilingual output is required.
Static analysis
No suspicious patterns detected.