T09 · Insecure Skill Coding Practices
Warning
- Location
SKILL.md:190- Finding
Non-Atomic Fixed-Window Rate Limiter Can Leave Counters Without Expiration
- Content
View full analysis
- Remediation
View remediation
Security audit
Redis缓存大师
Security checks for vulnerabilities and agentic risk
Overview
This Redis skill is a coherent production guide, but it enables live command execution and includes state-changing Redis guidance without clear confirmation or production safety boundaries.
Install only if you want an agent to advise on Redis operations, and require the agent to ask before running redis-cli or applying Redis config changes. Treat examples as patterns to review, not commands to run blindly against production; verify target host, database, key namespace, backups, and rollback plans first.
Vulnerability Patterns
- Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
- 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 (3)
T09 · Insecure Skill Coding Practices
Warning
- Location
SKILL.md:379- Finding
Sliding-Window Rate Limiting Is Not Enforced as a Single Atomic Operation
- Content
View full analysis
- Remediation
View remediation
tonumber(ARGV[4]) then return {0, count} end return {1, count} ``` The application must treat the script's returned decision as authoritative and avoid a second independent check. Validate the window, TTL, threshold, and member identifier before invocation. ]]>
T09 · Insecure Skill Coding Practices
Error
- Location
SKILL.md:171- Finding
Distributed Lock Lease Can Expire While the Original Owner Is Still Operating
- Content
View full analysis
- Remediation
View remediation
Vulnerability Patterns
- Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
- Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
- Behavioral ASTexec() Call, eval() Call, Dynamic Import
- Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
- Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (4)
Vague Triggers
Medium
- Category
- Not specified by scanner
- Confidence
- 94% confidence
- Finding
L017 将技能触发关键词定义为“Redis、缓存、TTL、淘汰策略、集群、分布式锁、限流、消息队列”等宽泛术语，其中多项是常见通用技术话题而非明确的技能调用条件。文档也未提供触发边界、限定上下文或负面示例，容易让与本技能并不相关的一般性提问触发该技能。
- Content
Missing User Warnings
Medium
- Category
- Not specified by scanner
- Confidence
- 92% confidence
- Finding
This section includes operational commands and configuration guidance such as changing persistence settings and runtime memory configuration that can directly alter a production Redis instance. Because the skill advertises exec capability and does not clearly require confirmation, environment checks, backup steps, or staging-first validation, an agent could apply these changes unsafely and cause service degradation or data loss.
- Content
Missing User Warnings
Medium
- Category
- Not specified by scanner
- Confidence
- 90% confidence
- Finding
The document recommends deletion/execution operations such as UNLINK/DEL-related guidance without prominently warning that these actions can permanently remove data or alter live system state. In an agentic context with exec enabled, concise examples may be copied or run automatically against production, increasing the chance of unintended destructive changes.
- Content
Natural-Language Policy Violations
Low
- Category
- Not specified by scanner
- Confidence
- 72% confidence
- Finding
L005-L006 使用中文显示名称与摘要，且全文默认以中文呈现，但文档未说明是否支持用户选择其他语言。对于通用技能，这种默认固定语言的自然语言设定可能与组织的语言/本地化选择要求不一致。
- Content
Static analysis
No suspicious patterns detected.