Back to skill

Security audit

deep-stock

Security checks for vulnerabilities and agentic risk

Overview

This is a disclosed stock-analysis prompt pack with no executable code or persistence, but it gives specific investment recommendations users should treat cautiously.

Install only if you want an AI-guided stock research framework. It may search current public market sources and produce buy, add, reduce, or clear-style recommendations, but it cannot guarantee accuracy and should not replace independent research or licensed financial advice.

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
  • 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 (5)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The example trigger phrases are very broad natural-language commands such as '分析比亚迪' / 'Analyze Xiaomi', which overlap with ordinary user requests and could cause the skill to activate unintentionally. This increases the chance of prompt routing confusion or unintentional invocation, especially in environments with multiple installed skills that respond to generic analysis requests.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The skill explicitly produces buy/hold/reduce/watch recommendations and presents them with confidence-style outputs, but it does not include a clear warning that the content is informational only and should not be treated as definitive financial advice. In a financial decision-making context, users may over-rely on the output and suffer material losses, especially because the skill markets itself as 'professional-grade' and action-oriented.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The skill provides explicit investment actions and position-sizing guidance such as buy/add/reduce/clear thresholds and recommended portfolio allocations, but does not include a clear warning that the output is not financial advice and may cause financial loss. In the context of a 'professional stock analysis framework,' users are likely to treat the recommendations as authoritative, increasing the chance of harmful reliance.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
77% confidence
Finding

Most of the skill instructions and examples are presented in Chinese, with English deferred to a later separate section. This can be a language/locale policy concern if users are not given an explicit choice of output language or informed that the skill may operate primarily in Chinese.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
84% confidence
Finding

The entire skill is written as a Chinese-only operating framework ('Charlie 的股票分析框架') and does not indicate that users may choose another language or locale. SQP-3 applies to all file types and flags language/locale policy violations when a skill forces a specific language without user opt-in.

Content

No source excerpt is available for this finding.