T09 · Insecure Skill Coding Practices
Warning
- Location
SKILL.md:14- Finding
Operational Portal Access Uses Unencrypted HTTP
- Content
View full analysis
- Remediation
View remediation
Security audit
承运商指数查询
Security checks for vulnerabilities and agentic risk
Overview
This skill is a straightforward carrier logistics analysis helper, but users should avoid using its plaintext HTTP portal links for sensitive operational data.
Install only if you intend the agent to read the specific spreadsheet you provide and use the listed Jingwe/JDL portal pages for carrier analysis. Prefer verified HTTPS access to the portal, confirm the destination before downloading anomaly details, and avoid exposing sensitive carrier or operational data over plaintext HTTP.
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 (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 (2)
Vague Triggers
Medium
- Category
- Not specified by scanner
- Confidence
- 91% confidence
- Finding
The activation condition is overly broad: any query about a carrier's logistics index score could trigger the skill without clearly constraining required inputs, scope, or safety checks. In a skill that reads local files and initiates web/download actions, ambiguous invocation increases the chance of unintended file access, accidental external navigation, or execution in the wrong user context.
- Content
Missing User Warnings
Medium
- Category
- Not specified by scanner
- Confidence
- 95% confidence
- Finding
The skill instructs the agent to read local Excel/CSV files, access external websites, and trigger downloads, but it does not present explicit user warnings, consent boundaries, or restrictions on what files/URLs may be touched. This creates a real risk of unauthorized local data exposure, unintended network access, and downloading sensitive operational details, especially because the skill also references anomaly-detail exports tied to carrier and date filters.
- Content
Static analysis
No suspicious patterns detected.