Back to skill

Security audit

qunar-travel-query

Security checks for vulnerabilities and agentic risk

Overview

The skill is a coherent Qunar travel-query helper, but its script can send the user's Qunar API key to any caller-supplied URL.

Install only if you trust every endpoint the agent will use. Before using this skill, restrict requests to known Qunar HTTPS domains and avoid supplying arbitrary or third-party URLs, because the current script can disclose the configured Qunar API key to the chosen endpoint.

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

Error
Location
scripts/qunar_query.py:71
Finding

Caller-Controlled API Endpoint Enables Credential Disclosure and Server-Side Request Forgery

Content
View full analysis
Remediation
View remediation
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (6)

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
88% confidence
Finding

The skill instructs the agent to use a script, configure credentials, and call external API endpoints, which implies environment access and outbound network use, yet it declares no explicit tool scope or permission boundaries. This is dangerous because the agent may be allowed to access credentials and make network requests without clear least-privilege constraints, increasing the risk of unintended data exposure or abuse if the skill or referenced resources are modified or adversarial.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
82% confidence
Finding

The skill metadata and user-facing documentation are entirely written in Chinese and present the skill as operating in that language, with no indication that users may choose another language or locale. Under the policy criteria, a language/locale constraint should either be optional or clearly justified; neither is stated here.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The document explicitly states that the skill supports custom API endpoints and parameters, which can cause user-supplied travel queries and other request data to be sent to arbitrary third-party services. In a travel-query skill, this increases SSRF/data-exfiltration style risk and privacy exposure, especially because users may assume requests only go to Qunar-controlled infrastructure.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
95% confidence
Finding

The external transmission here is not merely expected API usage because the destination is attacker-controlled via api_endpoint while the request includes the loaded API key in headers and arbitrary user-supplied payload data. In the context of a travel-query skill, outbound transmission should be limited to Qunar APIs, so sending secrets and request contents to any URL significantly increases the risk of credential leakage and misuse.

Content

Scanner excerpt · scripts/qunar_query.py (reported line 88)May include surrounding context.

python
timeout=30
            )
        else:
            response = requests.post(
                api_endpoint,
                headers=headers,
                json=api_params,

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
98% confidence
Finding

The script accepts a fully user-controlled --api_endpoint and then sends authenticated requests to that URL using the Qunar API key in the Authorization header. This enables server-side request forgery and credential exfiltration to arbitrary external hosts, because the secret intended for Qunar can be transmitted to attacker-controlled infrastructure.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
78% confidence
Finding

文件标题、说明、参数解释和帮助信息全部以中文呈现,没有提供用户可选择的语言或声明该技能仅面向特定中文用户群体。根据语言/区域策略,这可能构成未提供用户选择的语言限定。

Content

No source excerpt is available for this finding.