ClawHub
Back to skill

Security audit

HONGKONG-PAYMENT-QFPAY

Security checks for vulnerabilities and agentic risk

Overview

This is a Markdown-only QFPay payment API integration guide with sensitive but disclosed payment and refund examples.

Install only if you are intentionally integrating QFPay. Treat merchant app codes, keys, order IDs, logs, and production endpoints as sensitive; use sandbox or test credentials for examples and require explicit human approval before any live payment or refund action.

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
  • System Prompt LeakageDirect Leakage, Indirect Extraction, Tool-Based Exfiltration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Direct Prompt Extraction

High
Category
System Prompt Leakage
Confidence
85% confidence
Finding

Skill contains instructions that could directly expose system prompts, internal rules, or hidden instructions to users or external parties.

Content

Scanner excerpt · SKILL.md (reported line 571)May include surrounding context.

md
3. **Character Encoding**: All requests and responses use UTF-8 encoding.

4. **Timeout Handling**: For payment requests that don't return promptly, implement a polling mechanism to query transaction status.

5. **Async Notifications**: Configure `notify_url` to receive asynchronous payment completion notifications and verify notification signatures.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The README provides direct payment-order and status-query examples plus environment variable setup for merchant credentials, but it does not warn users about the sensitivity of API keys, the risk of running against production, or safe handling of real-money transactions. In a payment-processing skill, omission of these guardrails can lead to accidental live transactions, credential exposure, or unsafe testing practices by integrators.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The skill metadata and overview present the content as integration guidance, but the body includes executable examples that perform real payment, query, and refund calls against production-capable endpoints. This scope mismatch is dangerous because an agent or user may treat the skill as passive documentation while it actually enables state-changing financial operations, increasing the chance of unintended live transactions.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The refund example provides direct instructions for invoking a live refund endpoint, which exceeds the stated documentation-only scope of the skill. In a payment context, refund execution is a sensitive financial action; including runnable refund logic in an ostensibly guidance-focused skill raises the risk of unauthorized or accidental monetary reversals.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.