Security audit

DOGE节点免费版

Security checks for vulnerabilities and agentic risk

Overview

The skill is mainly for Dogecoin node and wallet RPC queries, but its broad/mismatched trigger text and under-disclosed wallet address generation make it a Review item.

Install only if you specifically want an agent to help with Dogecoin Core node and wallet RPC tasks. Do not rely on its current trigger text for generic database, SQL, analytics, or reporting work. Constrain use to explicit Dogecoin requests, review commands before execution, keep RPC bound to localhost, and require confirmation before generating new wallet addresses or exposing wallet balances and transaction history.

SkillSpector

Vulnerability Patterns

Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
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

Findings (6)

Description-Behavior Mismatch

High

Confidence: 98% confidence
Finding: The manifest description claims the skill should be used for database, SQL, and data storage tasks, while the actual content operates Dogecoin node and wallet RPC commands via exec. This mismatch can cause an agent to invoke a shell-capable blockchain management skill in unrelated contexts, increasing the chance of unintended command execution and exposure of wallet or node data.

Description-Behavior Mismatch

Medium

Confidence: 94% confidence
Finding: The trigger conditions say to use the skill for data analysis, reporting, statistics, and visualization, which does not match the documented Dogecoin-node functionality. Overbroad triggers can cause inappropriate invocation of an exec-enabled skill in contexts where it does not belong, expanding attack surface and confusing downstream automation.

Intent-Code Divergence

Medium

Confidence: 92% confidence
Finding: The documentation states the free version is read-only and does not involve write operations, but it explicitly includes getnewaddress, which mutates wallet state by creating a new address. This contradiction can bypass user expectations and policy checks that rely on the skill being non-mutating, making state-changing wallet actions appear safe.

Intent-Code Divergence

Medium

Confidence: 91% confidence
Finding: The FAQ says the free version only provides query capability, yet the skill elsewhere documents generating new wallet addresses. This inconsistent security model can mislead users and agents into approving operations under a false assumption that no wallet state can change.

Vague Triggers

High

Confidence: 96% confidence
Finding: The trigger description is overly broad and mismatched to the Dogecoin-node scope, making it more likely the skill will be auto-selected for unrelated requests. Because the skill has exec access, incorrect selection materially increases the risk of unnecessary shell execution and accidental interaction with sensitive local node or wallet resources.

Vague Triggers

High

Confidence: 98% confidence
Finding: The description includes vague invocation guidance for unrelated database and SQL tasks even though the skill actually performs Dogecoin CLI/RPC operations. Misclassification at the manifest level is dangerous in agent ecosystems because selection logic may route sensitive requests to a shell-executing tool with access to local blockchain and wallet context.

Static analysis

No suspicious patterns detected.