ClawHub
Back to skill

Security audit

sn-search-year-report

Security checks across malware telemetry and agentic risk

Overview

The skill is mostly a coherent annual-report search helper, but its bundled downloader can fetch any URL and overwrite any user-specified local file path without validation.

Review this skill before installing if your agent runs with broad filesystem or network access. It should be used only when you trust the requested URL and output path; ideally constrain downloads to known official report domains and a dedicated reports directory.

SkillSpector

Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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 (3)

Description-Behavior Mismatch

Medium
Confidence
93% confidence
Finding
The skill description says it searches annual-report sources, but the download subcommand can fetch any URL and write it locally. That expands the capability from search-only into arbitrary remote file retrieval, which can be abused to stage untrusted content on disk or access internal endpoints if the runtime has network reachability.

Context-Inappropriate Capability

Medium
Confidence
95% confidence
Finding
The download command is a generic URL-to-file primitive unrelated to the narrow annual-report search purpose. In an agent context, this broader capability increases risk of misuse for arbitrary content staging, SSRF-like internal fetches, or dropping attacker-controlled files into the workspace.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The subcommand writes attacker-influenced remote bytes directly to a user-specified path with no confirmation or path safety checks. In a tool-execution environment, this can overwrite files, place deceptive content in sensitive locations, or prepare follow-on attacks if other components later open or execute the written file.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.