Security audit

archive-free

Security checks for vulnerabilities and agentic risk

Overview

The skill is mostly a local content archiver, but it asks for broad command execution and advertises unrelated use cases that users should review before installing.

Install only if you want an agent to save external articles, tweets, PDFs, and video metadata as persistent local Markdown files. Review the broad exec permission and unrelated activation claims first; prefer a version that removes generic command execution or clearly limits it to archive setup commands.

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: 95% confidence
Finding: The manifest markets this as an archive/snapshot skill but also advertises unrelated media processing, audio editing, dubbing, and SEO use cases. This kind of scope inflation is dangerous because it can cause agents or users to invoke the skill in contexts far beyond its documented behavior, increasing the chance of unsafe tool use and policy bypass through misleading capability claims.

Description-Behavior Mismatch

Medium

Confidence: 91% confidence
Finding: Later sections broaden the skill into generic file handling, API integration, and command execution even though the core purpose is content archiving. That mismatch can mislead an agent into treating a narrow archival tool as a general automation primitive, which expands attack surface and weakens trust in the manifest as a security boundary.

Context-Inappropriate Capability

Medium

Confidence: 97% confidence
Finding: The manifest grants exec alongside read and write for a content archiving skill without a clear, necessary justification. Unnecessary command execution is a serious privilege expansion because any prompt-driven misuse, spec ambiguity, or future implementation can turn a simple archive request into local command execution with file-system side effects.

Context-Inappropriate Capability

Medium

Confidence: 95% confidence
Finding: The documentation openly promotes command execution as a user-facing capability despite the skill's stated purpose being archival and search. Normalizing exec in the public interface increases the likelihood that users or agents will route arbitrary operational tasks through this skill, creating a larger abuse surface for command injection or unintended local actions.

Vague Triggers

High

Confidence: 92% confidence
Finding: The broad 'Use when' language spans unrelated domains such as media tooling and SEO, making the activation criteria ambiguous and overly permissive. In agent ecosystems, vague routing language is dangerous because it can trigger the skill in inappropriate contexts where its tools and persistence behavior are not expected by the user.

Missing User Warnings

Medium

Confidence: 84% confidence
Finding: The skill describes capturing external content as local snapshots, but the point-of-use description does not clearly foreground that files will be created and persisted locally. This is risky because users may unintentionally store sensitive, regulated, or copyrighted material on disk without realizing the retention implications.

Static analysis

No suspicious patterns detected.