ClawHub
Back to skill

Security audit

Timer

Security checks for vulnerabilities and agentic risk

Overview

The skill is mostly a timer/reminder guide, but it asks for broad execution/write authority and contains unrelated workflow, API, callback, and file-processing claims that are not well scoped to timers.

Install only if you want a session-bound background timer skill and are comfortable allowing it to start and manage background commands. Avoid putting secrets in reminder text, keep timer counts reasonable, and treat the unrelated API, callback, file-processing, and workflow claims as under-scoped documentation rather than validated features.

SkillSpector

Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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 (3)

Vague Triggers

High
Confidence
94% confidence
Finding
The skill description tells the agent to use this timer skill for unrelated project management, task planning, progress tracking, team collaboration, and workflow scenarios. Overly broad invocation guidance can cause the agent to select a tool with `exec`, `write`, and background-process capabilities in contexts where it is unnecessary, increasing the chance of unintended command execution or misuse of privileged tooling.

Natural-Language Policy Violations

Medium
Confidence
90% confidence
Finding
The metadata states that reminder content 'must' be conveyed to the user without considering the user's language or locale. In an agent setting, hard-coded delivery requirements can override normal conversational safeguards and produce misleading or inaccessible notifications, especially for multilingual users or contexts requiring consent and formatting control.

Natural-Language Policy Violations

Medium
Confidence
91% confidence
Finding
The reply contract mandates fixed output behavior and explicitly forbids certain prefixes to avoid filtering, but it does not require matching the user's language. This can pressure the agent to prioritize tool-specific formatting rules over user comprehension and safe interaction norms, making the behavior more brittle and easier to misuse in multilingual contexts.

Static analysis

No suspicious patterns detected.