ClawHub
Back to skill

Security audit

Qwen

Security checks for vulnerabilities and agentic risk

Overview

This Qwen skill is a documentation-only helper that clearly describes its cloud API use and only stores local notes after user approval.

Install only if you are comfortable using Qwen workflows. Hosted mode sends prompts, and possibly images, to Alibaba Cloud Model Studio, so use self-hosted routes or redact sensitive data for confidential work. Do not put API keys into the ~/qwen/ notes; keep them in environment variables as the skill instructs.

A.I.G

Vulnerability Patterns
  • 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
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (5)

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
75% confidence
Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · SKILL.md (reported line 127)May include surrounding context.

md
- stores lightweight local notes only after user approval

This skill NEVER:
- invent live model availability without checking
- persist secrets in `~/qwen/`
- execute destructive downstream automation without validated output
- pretend one backend's tool-calling behavior applies everywhere

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

The hosted examples instruct users to send prompts and ticket-like content to third-party cloud endpoints but provide no warning about privacy, data handling, or the need to avoid sensitive data. In a skill intended to help build routing workflows, this omission can lead to accidental transmission of proprietary or personal information to remote infrastructure.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
85% confidence
Finding

This example sends user-supplied content to a remote hosted chat completion API over the network. While the transmission is intentional, the security issue is the lack of guardrails or disclosure around external data sharing, which can cause users to expose sensitive internal content during routine use.

Content

Scanner excerpt · api-patterns.md (reported line 29)May include surrounding context.

Minimal Hosted Chat Completion

bash
curl -s "${QWEN_BASE_URL}/chat/completions" \
  -H "Authorization: Bearer $DASHSCOPE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{

External Transmission

Medium
Category
Data Exfiltration
Confidence
90% confidence
Finding

The strict JSON pass example transmits a support-ticket style prompt to a hosted remote endpoint, increasing the likelihood that real incident, account, or customer data could be copied into the request. Without privacy warnings or redaction guidance, users may unintentionally leak sensitive operational information.

Content

Scanner excerpt · api-patterns.md (reported line 47)May include surrounding context.

Use a separate normalization pass when the first answer needs reasoning freedom:

bash
curl -s "${QWEN_BASE_URL}/chat/completions" \
  -H "Authorization: Bearer $DASHSCOPE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · api-patterns.md (reported line 78)May include surrounding context.

Minimal Local Chat Check

bash
curl -s "${LOCAL_BASE_URL}/chat/completions" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "MODEL_ID",