ClawHub
Back to skill

Security audit

OmniFocus MCP for OpenClaw

Security checks for vulnerabilities and agentic risk

Overview

This skill openly lets OpenClaw read and update OmniFocus tasks on a Mac, and its behavior matches that stated purpose.

Install only if you want OpenClaw to access and change your local OmniFocus data. Start with read-only queries, review proposed bulk edits or deletions carefully, and remember that macOS automation permission lets the host process operate OmniFocus on your behalf.

SkillSpector

Vulnerability Patterns
  • 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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Vague Triggers

Medium
Confidence
88% confidence
Finding
The skill description is broadly phrased around reading or updating OmniFocus from natural language, which can cause over-activation for loosely related planning or task-management requests. Because this skill supports write and delete operations against a real task database, ambiguous invocation increases the chance of unintended tool use and accidental modifications.

Vague Triggers

Medium
Confidence
81% confidence
Finding
The default prompt encourages broad task-management operations such as reviewing, triaging, creating, and reorganizing OmniFocus data without clearly constraining scope, confirmation requirements, or read-only defaults. In an agent setting, this can lead to unintended state-changing actions being invoked implicitly, especially because the skill is positioned for natural-language use and the policy allows implicit invocation.

Static analysis

No suspicious patterns detected.