ClawHub
Back to skill

Security audit

cross-machine-offline-taskbox

Security checks for vulnerabilities and agentic risk

Overview

This skill is a local offline task queue with an Ollama worker; its local file and subprocess behavior is purpose-aligned, though users should note the broad triggers and local task persistence.

Install only if you want a local offline task queue and are comfortable storing task titles, details, and results in a local taskbox.json file. Review task contents before exporting sync packs because exports contain all tasks, and run the offline worker only when you intend queued task text to be processed by your local Ollama model.

SkillSpector

Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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)

Lp3

Medium
Category
MCP Least Privilege
Confidence
90% confidence
Finding
The skill metadata declares no permissions, yet the analyzer detected capabilities associated with file read, file write, and shell access. This creates a transparency and least-privilege problem: users and hosting platforms may invoke a skill that can affect the local system more powerfully than its manifest suggests, especially dangerous here because the skill is explicitly positioned around offline, cross-machine, and local-model task execution.

Vague Triggers

Medium
Confidence
78% confidence
Finding
The trigger keywords 'cross', 'machine', 'offline', and 'taskbox' are broad and generic, making accidental activation more likely in unrelated conversations. Because this skill appears to have elevated operational capabilities, over-broad triggering increases the chance that a sensitive or system-affecting workflow is engaged without clear user intent.

Vague Triggers

Medium
Confidence
90% confidence
Finding
The manifest advertises many broad trigger phrases such as '省积分', '批处理', 'Ollama', and 'local LLM' that can plausibly appear in ordinary user conversations. This can cause the skill to activate outside a clearly bounded use case, increasing the chance of unintended routing, unexpected handling of sensitive task descriptions, or abuse through prompt steering into the skill.

Static analysis

No suspicious patterns detected.