Back to skill

Security audit

投资五问决策框架

Security checks for vulnerabilities and agentic risk

Overview

This is a markdown-only Chinese investment analysis framework with no executable payload, persistence, or hidden data access, but users should treat its investment conclusions cautiously.

Before installing, understand that this skill may generate concrete investment labels such as buy, watch, or abandon. Verify all market data, dates, and calculations independently, and do not rely on it as personalized financial advice or as a substitute for a qualified advisor.

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
Vulnerability Patterns
  • 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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (5)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The README gives explicit investment decision guidance such as '买入', '观察', and '放弃', and presents threshold-based rules that could be relied on as actionable financial advice. Because there is no risk disclosure, suitability warning, or statement that outputs are educational only and may be inaccurate or outdated, users may over-trust the skill and make harmful financial decisions.

Content

No source excerpt is available for this finding.

Skill Enumeration

Medium
Category
Agent Snooping
Confidence
80% confidence
Finding

The README discloses an internal local filesystem path to the skill under the user's home directory. While not highly sensitive on its own, this is environment-enumeration information that can help an attacker understand the agent's directory structure and craft more targeted prompts or path-based attacks in a broader exploit chain.

Content

Scanner excerpt · README.md (reported line 104)May include surrounding context.

md
---

## 技能位置
`~/.openclaw/workspace/skills/investment-five-questions/SKILL.md`

---

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The skill explicitly produces buy/watch/abandon investment recommendations and even suggests querying market data, but it does not prominently warn users that outputs are informational only and not financial advice. In this context, users may over-rely on the model’s recommendations for real asset allocation, creating foreseeable financial harm from inaccurate, stale, or oversimplified analysis.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
88% confidence
Finding

All headings, examples, and invocation patterns are written in Chinese, and the example interactions assume Chinese-only use without any note that other languages are supported. Under the language/locale policy, a skill should not implicitly require a specific language unless the constraint is documented and justified or the user is offered a choice.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
95% confidence
Finding

The manifest-style metadata specifies 'language: zh-CN', which indicates a fixed language/locale. Under the policy, a forced language or locale should be flagged unless the skill offers user choice or clearly justifies a region-specific constraint, which is not present here.

Content

No source excerpt is available for this finding.