ClawHub
Back to skill

Security audit

okx-sentiment-tracker

Security checks for vulnerabilities and agentic risk

Overview

The skill is mainly a read-only OKX news and sentiment helper, but it also contains under-disclosed workflows that can use account positions and balances for personalized trading guidance.

Install only if you are comfortable configuring OKX live API credentials for a news and sentiment tool. Before using it, verify your OKX API key permissions are read-only and be cautious with any workflow that asks to inspect positions, balances, leverage, or P&L; that behavior goes beyond ordinary news lookup and should require explicit user intent.

SkillSpector

Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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
Findings (4)

Context-Inappropriate Capability

Medium
Confidence
89% confidence
Finding
The skill explicitly instructs the agent to use general web search as a fallback, which extends behavior beyond the declared OKX-only scope and can cause the agent to retrieve untrusted external content. That broadening increases attack surface for prompt injection, source spoofing, and inconsistent trust boundaries, especially when users expect answers grounded only in the OKX toolchain.

Context-Inappropriate Capability

Medium
Confidence
92% confidence
Finding
This workflow extends a read-only news/sentiment skill into portfolio-aware position impact analysis by instructing the agent to access account positions and balance, then produce actionable trading guidance. Even though it warns not to place orders without confirmation, it still expands scope into sensitive financial data handling and personalized investment advice, which increases privacy, compliance, and user-harm risk if invoked from this skill context.

Vague Triggers

Medium
Confidence
93% confidence
Finding
The description contains very broad triggers such as 'any crypto news', 'latest news', 'market update', 'daily briefing', and catch-all phrasing for 'any request for crypto news, market intelligence, sentiment analysis, social trend data, or macro-economic calendar.' This can cause the skill to activate for loosely related requests, increasing the chance of unintended tool use, unnecessary credentialed API access, and incorrect routing away from safer or more appropriate skills.

Vague Triggers

Medium
Confidence
90% confidence
Finding
The manifest description contains very broad trigger phrases such as 'latest news', 'market update', 'full article', and generic macroeconomic terms, which can cause the orchestration layer to invoke this skill for many unrelated or only partially related user requests. Because the skill requires API credentials and operates on external news/sentiment data, overbroad matching increases the chance of unintended tool use, unnecessary credential exposure to the skill runtime, and incorrect routing away from more appropriate skills.

Static analysis

No suspicious patterns detected.