ClawHub
Back to skill

Security audit

BSC Analytics BSC生态分析

Security checks for vulnerabilities and agentic risk

Overview

This skill is a BSC analytics package, but it overstates live monitoring capabilities and includes an under-disclosed paid verification flow with an exposed API key.

Review before installing. Treat its analytics and yield output as static sample research, not live BSC monitoring or financial advice. Avoid providing a wallet address unless you intend to use the paid SkillPay verification flow, and the publisher should rotate the exposed key, move payment verification server-side or to secure configuration, disclose the external payment data flow, and clarify which monitoring features are actually implemented.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • 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
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Context-Inappropriate Capability

Medium
Confidence
91% confidence
Finding
The environment variable SKILLPAY_SKIP_VERIFICATION allows payment checks to be bypassed entirely, creating a trivial authorization bypass if this code runs in any environment where users or adjacent systems can influence environment variables. Even if intended for testing, leaving this path in production code weakens access control around paid functionality.

Vague Triggers

Medium
Confidence
81% confidence
Finding
The trigger description is very broad and can cause the skill to activate for many generic BSC-related requests, including ones that may not require this skill's scripts, references, or investment-oriented guidance. Over-broad activation increases the chance of unnecessary tool use, irrelevant network/script execution in downstream implementations, and unprompted financial analysis that could mislead users.

Vague Triggers

Low
Confidence
77% confidence
Finding
The example phrases are conversational and expansive, such as asking for good mining opportunities or general ecosystem status, without defining scope limits or requiring user confirmation before producing investment-oriented output. In a crypto-investment context, this can steer the skill toward giving broad, potentially risky recommendations when the user intent is ambiguous.

Missing User Warnings

High
Confidence
99% confidence
Finding
A live API credential is hardcoded directly in source code and then used in outbound requests. If the skill code is exposed, reused, logged, or committed to a repository, attackers can extract the key and abuse the payment API, impersonate the skill, or consume/modify billable verification operations.

Static analysis

No suspicious patterns detected.