Security audit

自动化配方

Security checks across malware telemetry and agentic risk

Overview

This automation skill is not malicious, but it asks for broad file and command authority while documenting delete/rename/sync and vague external API use without clear scoping or confirmation rules.

Install only if you are comfortable with a general-purpose automation helper that may read, write, rename, sync, and potentially delete files. Use it with explicit folders and dry-run or backup steps, and do not provide API keys or allow networked workflows unless the exact service, data sent, and purpose are clear.

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
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands

Findings (6)

Description-Behavior Mismatch

Medium

Confidence: 87% confidence
Finding: The skill is presented as a local automation recipe pack, but later documentation expands its scope to external API integration, network access, and command execution. This mismatch can cause an agent or user to grant broader trust and permissions than expected, increasing the chance of unintended data exfiltration or remote interactions under a benign-looking label.

Context-Inappropriate Capability

Medium

Confidence: 84% confidence
Finding: The documentation introduces API-key handling and external service access without a clear, necessary link to the advertised automation-recipes purpose. In an agent environment with file and shell access, ambiguous justification for network/API usage raises the risk that sensitive local data could be transmitted externally without sufficiently informed consent.

Vague Triggers

Medium

Confidence: 79% confidence
Finding: The description says to use the skill whenever users want efficiency improvements, automation, batch processing, or workflow optimization, which is extremely broad. Overly generic routing criteria can cause the agent to invoke a file- and shell-capable skill in many situations where a narrower, safer skill or plain reasoning would suffice.

Vague Triggers

Medium

Confidence: 81% confidence
Finding: The instruction to directly call the skill from an AI Agent conversation based on a general task description lacks meaningful scope constraints. In practice, this can lead to automatic invocation on underspecified requests, increasing the chance of unreviewed file modifications, command execution, or risky workflow actions.

Missing User Warnings

Medium

Confidence: 90% confidence
Finding: The skill advertises duplicate-file cleanup, directory synchronization with deletion strategies, and batch renaming, but does not prominently warn that these actions can modify or delete user files. For a skill with Write, Edit, and Bash tools, missing upfront warnings materially increases the risk of accidental destructive operations.

Missing User Warnings

Low

Confidence: 83% confidence
Finding: The documentation mentions external services, API keys, and network connectivity, but does not give a clear user-facing warning that the skill may transmit data off-host or require secrets. This omission weakens informed consent and can cause users to expose credentials or sensitive content unexpectedly.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.