ClawHub
Back to skill

Security audit

Spotlight Search

Security checks across malware telemetry and agentic risk

Overview

This is a coherent macOS Spotlight search helper with expected local file-search access and a disclosed admin reindex command that requires explicit user approval.

Install only if you are comfortable letting the agent search your Mac's Spotlight index and show local file paths or metadata in chat. Prefer folder-scoped searches for sensitive areas, and do not approve sudo mdutil reindexing unless you intentionally want a system-wide Spotlight rebuild.

SkillSpector

Vulnerability Patterns
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Vague Triggers

Medium
Confidence
85% confidence
Finding
The trigger phrases are broad enough that the skill could be invoked unintentionally for generic requests like 'find this file on mac' or 'search my computer.' While the skill itself is not overtly dangerous, over-broad auto-invocation can cause unnecessary filesystem searches, surface sensitive file paths or metadata, and increase exposure when used in contexts the user did not explicitly intend.

Sudo/Root Execution

Medium
Category
Privilege Escalation
Content
1. Check indexing status: `mdutil -s /`
2. Ensure the directory is indexed (Spotlight excludes some system and Library folders)
3. Re-index only with explicit user approval because it is a privileged, system-wide action: `sudo mdutil -E /`

**Query syntax errors?**
Confidence
78% confidence
Finding
The skill includes a `sudo mdutil -E /` command that performs a privileged, system-wide reindex of Spotlight. Even though the text says to do this only with explicit user approval, embedding a root-level command in an agent skill creates risk of privilege escalation prompts, disruptive system changes, and unintended modification of indexing state if invoked carelessly or by an over-permissive agent.

VirusTotal

60/60 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.