Back to skill

Security audit

Move

Security checks for vulnerabilities and agentic risk

Overview

The skill is not destructive, but it is advertised as a Move blockchain reference while its actual content is generic finance and checklist boilerplate.

Review this skill carefully before installing. It appears low-risk to run locally, but it should not be relied on for Move programming, blockchain audits, or financial/compliance decisions unless the publisher corrects the documentation and replaces the generic placeholder content with accurate domain-specific material.

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
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • 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)

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

The skill is presented as a Move blockchain reference tool, but its commands and descriptions map to generic finance, compliance, and investment-style topics rather than Move-specific content. This kind of capability misrepresentation can mislead users or downstream agents into invoking the skill in inappropriate contexts, increasing the risk of bad guidance, compliance errors, or trust abuse.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
Not specified by scanner
Confidence
98% confidence
Finding

The documented purpose and the exposed command set are materially inconsistent: a purported Move blockchain skill instead offers broad finance/regulatory/risk content. In an agent ecosystem, this can cause incorrect tool selection and unsafe reliance on irrelevant or misleading advice, especially in blockchain or financial workflows where precision matters.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

High
Category
Not specified by scanner
Confidence
97% confidence
Finding

The documentation contradicts itself by advertising Move/blockchain usage while command summaries describe finance subject matter such as regulations, instruments, and strategies. This contradiction is dangerous because it undermines operator trust and can steer users or autonomous agents toward decisions based on the wrong domain assumptions.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The script header claims the tool is for Move in blockchain contexts, but the implementation provides nonspecific boilerplate text unrelated to that domain. In an agent skill ecosystem, deceptive or misleading capability descriptions can cause agents to select the wrong tool and propagate inaccurate output, which is especially risky where blockchain actions, audits, or code guidance may be inferred from the skill name.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The skill is labeled as a Move blockchain reference tool, but the exposed commands return generic finance, risk, and compliance placeholder content instead of Move- or blockchain-specific guidance. This is dangerous because users and downstream agents may rely on incorrect domain assumptions, leading to misuse, bad decisions, or unsafe automation in a security-sensitive blockchain context.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.