Back to skill

Security audit

租车自驾行前助手:准备清单 + 各国注意事项 + 经典路线

Security checks for vulnerabilities and agentic risk

Overview

This skill is a coherent Chinese car-rental road-trip planning template with limited local impact, though generated HTML should be handled carefully.

Installers should understand this skill is mainly a Chinese HTML template for rental-car trip planning. If generated pages are shared or hosted, ensure user-supplied values are HTML-escaped or sanitized first, and confirm language expectations for non-Chinese users.

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)

T09 · Insecure Skill Coding Practices

Warning
Location
html-template.md:157
Finding

Unescaped User-Controlled Values in Generated HTML

Content
View full analysis

🚗 {location} · {days}天租车自驾行前计划

{experience_level_label} · {trip_type_label} · {drivers_label} · {style_preference_label}
``` Other direct interpolation locations include: ```html {location} · {days}天租车自驾行前计划 ``` ```html
📋 {location} 租车关键注意事项
``` The template requires direct placeholder replacement without establishing an output-encoding requirement: ```text - All `{placeholders}` must be replaced with actual content during generation; the final HTML should not contain brace-delimited placeholders. ``` ### Technical Analysis The Skill instructs the agent to place user-derived values, particularly the unrestricted `location` string, directly into an HTML document. It does not require HTML escaping, sanitization, length limits, or rejection of markup before interpolation. If a value containing HTML syntax is copied into the template verbatim, it can terminate the intended text context and introduce attacker-controlled elements. For example, a malicious location value could close the existing element and insert a `
Remediation
View remediation
`, `"`, and `'`. 2. Explicitly prohibit treating user-supplied or generated textual values as raw HTML. 3. Validate `days` as a bounded integer and validate enum fields against their documented allowlists before rendering. 4. Constrain `location` with a reasonable maximum length and reject control characters or markup when markup is not required. 5. Use a templating system with automatic escaping enabled instead of unrestricted textual placeholder replacement. 6. Keep dynamic values in text-node contexts. If values are ever inserted into attributes, URLs, CSS, or JavaScript, apply encoding specific to those contexts. 7. Add tests using payloads such as `
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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 (3)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The trigger list includes broad, common travel phrases such as '自驾攻略' and '租车流程', which can cause the skill to activate for general travel-planning requests that do not specifically need this specialized HTML-output workflow. Over-broad invocation increases the chance of unintended routing, user confusion, and policy or UX mismatches, especially because the skill imposes a fixed output format and domain assumptions.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill description and body require Chinese-language output for all users without checking the user's language preference, which can override user intent and reduce usability or accessibility for non-Chinese speakers. In an agent setting, forced language output can also cause incorrect skill selection or failure to satisfy the original request, especially when triggers include English phrases like 'road trip' and 'car rental checklist'.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The HTML template hard-codes lang="zh-CN", which enforces a specific language/locale for generated output. The file does not indicate that users can choose another language or that the Chinese locale is an optional, justified regional constraint.

Content

No source excerpt is available for this finding.