ClawHub
Back to skill

Security audit

Obsidian CLI(免费版)

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed Obsidian command-line helper for reading, searching, and editing a local vault, with no hidden code or exfiltration behavior found.

Install only if you want an agent to operate on your local Obsidian vault. Ask the agent to use this skill only for explicit Obsidian tasks, and require confirmation before commands that create notes, append content, edit frontmatter properties, or toggle tasks, especially in a primary vault without backups or version control.

SkillSpector

Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Vague Triggers

Medium
Confidence
88% confidence
Finding
The skill’s activation scope is overly broad, combining generic phrases like project management, planning, and collaboration with Obsidian-specific terms. This can cause an agent to invoke the skill in contexts where the user did not clearly intend filesystem-backed note operations, increasing the chance of unintended reads or writes to a local knowledge base.

Missing User Warnings

Medium
Confidence
93% confidence
Finding
The skill describes creation and modification capabilities but does not clearly warn that commands can change local files in the user’s vault. In an agent setting, omission of mutation warnings can lead to silent data changes, overwrites, or note corruption when the user expected only retrieval or analysis.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The examples include append, prepend, property:set, property:remove, task toggle, and task done operations without any explicit warning that these actions alter note contents and metadata. Example-heavy skills are often copied verbatim by agents, so presenting mutating commands without guardrails materially increases the risk of unintended, irreversible edits.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.