ClawHub
Back to skill

Security audit

进度任务管理

Security checks across malware telemetry and agentic risk

Overview

This skill is a straightforward local task tracker that writes task and progress notes in the workspace.

Install this if you want the agent to maintain a local task board and daily progress log. Avoid putting secrets or sensitive personal details into task text, and periodically review TASKS.md, memory files, and .learnings for unwanted or inaccurate entries.

SkillSpector

By NVIDIA
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 (3)

Vague Triggers

Medium
Confidence
93% confidence
Finding
The skill is configured with `always: true` and broad natural-language triggers, so it may activate on routine conversation and modify workspace files without sufficiently explicit user intent. In a file-writing skill, unintended activation is risky because it can create or alter `TASKS.md` and memory logs, causing integrity issues and surprise persistence of user content.

Vague Triggers

Medium
Confidence
95% confidence
Finding
The trigger phrase `给我一个明确的任务` is semantically vague and can match many ordinary user requests that are not intended to create or mutate tracked tasks. Because this skill performs writes and state transitions, ambiguity increases the chance of unauthorized or accidental task creation and persistent record changes.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The skill describes automatic updates to `TASKS.md` and `memory/今天.md` but does not prominently warn users that ordinary task-related requests may cause persistent workspace writes. Lack of disclosure is dangerous here because users may expect a conversational summary, not file mutation, leading to unintended persistence of sensitive or noisy content.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.