Back to skill

Security audit

基于艾宾浩斯遗忘曲线的英语六级词汇记忆系统

Security checks for vulnerabilities and agentic risk

Overview

The skill is a CET-6 study helper, but its reminder setup uses persistent cron jobs that run an unaudited root-path CET-4 script and send messages through a default account.

Install only after reviewing the reminder behavior. Require explicit confirmation before any cron change, verify or replace the referenced reminder script, avoid root-owned paths and the default messaging account, and make sure there is a clear way to list, disable, and delete stored schedule data and reminders.

Vulnerability Patterns
  • System PersistenceInstalls backdoors, hooks, services, or scheduled tasks that survive the run
  • 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)

T06 · System Persistence

Error
Location
references/schedule-config.md:11
Finding

Persistent Cron Jobs Execute an Unbundled External Reminder Script

Content
View full analysis
\ --message "" ``` ### Technical Analysis The Skill directs the Agent to modify the user's `crontab`, creating scheduled tasks that survive the current Skill invocation and execute every day. This constitutes system persistence. Recurring reminders are part of the declared functionality, so scheduling has a legitimate product purpose. However, direct modification of system cron exceeds the minimum privilege necessary when an application-scoped or Agent-managed scheduler could provide the same feature without establishing operating-system-level persistence. The cron entries execute `/root/.openclaw/scripts/cet4-reminder.sh`, but that script is not included in the audited project. Its behavior and integrity therefore cannot be verified. Any process able to create or replace the referenced script could cause cron to execute different commands automatically at the configured times. The use of a `/root` path also suggests execution in a highly privileged environment, although the project does not establish which account ultimately owns the cron entries. There is also a CET-4/CET-6 mismatch: the Skill declares CET-6 functionality while the cron configuration searches for and executes a CET-4 reminder. This can cause incorrect duplicate detection, accidental reuse of unrelated tasks, or execution of a script belonging to another Sk ...[truncated 1515 chars]
Remediation
View remediation
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Rogue AgentSelf-Modification, Session Persistence
  • YARA SignaturesMalware Match, Webshell Match, Cryptominer Match
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (14)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The skill instructs the agent to write persistent user data to CSV files and set up a scheduled task, but it does not require explicit user notice or consent about persistence and system changes. This can lead to unexpected data retention and background automation, which is a real security and privacy issue even in an educational context because the actions outlive the immediate conversation.

Content

No source excerpt is available for this finding.

Ae4

Medium
Category
analysis-evasion
Confidence
80% confidence
Finding

Suspicious Unicode normalization or mixed-script content

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
98% confidence
Finding

The title and introduction describe this as a CET-6 complete vocabulary list, including roughly 5,500 core words. However, the summary at L4339 states '约 4500 个四级核心词汇', which directly contradicts the earlier CET-6 framing and changes the skill's documented educational scope.

Content

No source excerpt is available for this finding.

YARA rule 'network_reconnaissance': Network reconnaissance and scanning patterns [hacktools]

Medium
Category
YARA Match
Confidence
65% confidence
Finding

YARA rule matched a hack tool or exploit indicator (offensive tools, reconnaissance, privilege escalation, or exploit frameworks).

Content

Scanner excerpt · references/cet6-vocabulary.md (reported line 1349)May include surrounding context.

md
| 同伴 |
| female | /ˈfiːmeɪl/ | adj./n. | 女性的 |
| fence | /fens/ | n. | 栅栏 |
| ferry | /ˈferi/ | n. | 渡船 |
| festival | /ˈfestɪvl/ | n. | 节日 |
| fetch | /fetʃ/ | v. | 取来 |
| fever | /ˈfiːvər/ | n. | 发烧 |
| few | /fjuː/ | adj./n. | 很少的 |
| fiber | /ˈfaɪbər/ | n. | 纤维 |
| fiction | /ˈfɪkʃn/ | n. | 小说 |
| field | /fiːld/ | n. | 田地 |
| fierce | /fɪəs/ | adj. | 凶猛的 |
| fifteen | /ˌfɪfˈtiːn/ | num. | 十五 |
| fifth | /fɪfθ/ | num. | 第五 |
| fifty | /ˈfɪfti/ | num. | 五十 |
| fight | /faɪt/ | v./n. | 战斗 |
| figure | /ˈfɪɡər/ | n. | 数字;人物 |
| file | /faɪl/ | n. | 文件 |
| fill | /fɪl/ | v. | 填满 |
| film | /fɪlm/ | n. | 电影 |
| filter | /ˈfɪltər/ | n./v. | 过滤 |
| final | /ˈfaɪnl/ | adj./n. | 最后的 |
| finally | /ˈfaɪnəli/ | adv. | 最终 |
| finance | /ˈfaɪnæns/ | n. | 金融 |
| financial | /faɪˈnænʃl/ | adj. | 金融的 |
| find | /faɪnd/ | v. | 找�

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

This markdown file presents all instructions and labels exclusively in Chinese, with no note that the skill is China/Chinese-specific and no option for other user languages. That can violate language/locale policy when users are not given an explicit language choice or opt-in.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The instructions direct persistent modification of system scheduling via crontab, but do not require an explicit user-facing warning or informed consent about installing recurring background tasks. This is dangerous because it creates durable behavior on the host that may surprise users, persist beyond the session, and be abused to normalize silent persistence mechanisms.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

技能清单声明这是 CET-6 词汇记忆系统,但该文档中的 crontab 示例调用的是 cet4-reminder.sh,后续检查命令也匹配 cet4。这不是单纯的信息缺失,而是文档明确指向另一类考试词汇提醒实现,和技能宣称用途直接矛盾。

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
98% confidence
Finding

L27 的注释明确写着“CET-4 单词记忆提醒脚本”,而技能 manifest 说明该技能服务于 CET-6 备考。该注释会误导开发者或维护者认为此技能操作的是 CET-4 数据与提醒流程,属于文档意图与技能实际宣称范围的直接冲突。

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
84% confidence
Finding

The reminder script template sends automated messages to a user account without any notice about what data is used, where it is sent, or how often transmission occurs. This is risky because background messaging can expose identifiers and behavioral data, and recurring outbound messages can occur without meaningful user awareness.

Content

No source excerpt is available for this finding.

Session Persistence

Medium
Category
Rogue Agent
Confidence
87% confidence
Finding

Enumerating and relying on existing crontab entries indicates interaction with persistent host-level state, which can reveal prior scheduled tasks and facilitate modification of recurring execution. In this skill's context, that increases risk because the feature is specifically designed to establish long-lived reminders and lacks strong disclosure and isolation guidance.

Content

Scanner excerpt · references/schedule-config.md (reported line 43)May include surrounding context.

设置前检查用户是否已有定时任务:

bash
crontab -l | grep cet4

如果没有,则添加;如果有,确认时间是否与用户设置一致。

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
92% confidence
Finding

This markdown file presents the content entirely in Chinese from the title onward, with no indication that users can opt into another language or that the file is intentionally limited to a Chinese-speaking audience. Under the language/locale policy check, a forced language choice without opt-in can be a natural-language policy concern.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Low
Category
Not specified by scanner
Confidence
99% confidence
Finding

A section header for the letter V implies the following entries should be V-initial vocabulary, but the content starts by duplicating U-section words such as 'ugly', 'ultimate', and 'unable'. This is an active contradiction between the section documentation and the actual listed content.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
88% confidence
Finding

The file specifies Chinese-language setup text and a Chinese reminder message template, but does not indicate that language is configurable or intentionally limited to a Chinese-speaking context. This can violate language/locale policy when users are not offered a language choice or opt-in.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
97% confidence
Finding

This markdown file presents all schema documentation in Chinese, including headings and field descriptions, with no indication that the skill supports other languages or that Chinese is a required locale. Under the policy, forcing a specific language without user opt-in can be a natural-language policy violation.

Content

No source excerpt is available for this finding.