ClawHub
Back to skill

Security audit

Work Productivity Gog Google Workflow Helper

Security checks for vulnerabilities and agentic risk

Overview

This is a documentation-only workflow helper with no executable code or persistence, though its activation wording is broad and could cause accidental invocation.

Before installing, be aware that this skill may activate for general Google Workspace, CLI, Gmail, Drive, Contacts, or bug-fix requests because implicit invocation is enabled and the trigger terms are broad. It appears safe as a documentation-only helper, but users should invoke it explicitly for Gog-style workflow planning and avoid relying on fuzzy routing for sensitive workspace tasks.

A.I.G

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

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 (11)

Vague Triggers

High
Category
Not specified by scanner
Confidence
97% confidence
Finding

The trigger phrases are broad enough to match ordinary productivity, Google Workspace, CLI, or bug-fix requests that may not specifically intend to invoke this skill. In an agent-routing context, this can cause over-triggering, unintended skill selection, and execution of workflows outside the user's intended scope, which is a real security and reliability concern because it expands the skill's effective authority surface.

Content

No source excerpt is available for this finding.

Vague Triggers

High
Category
Not specified by scanner
Confidence
97% confidence
Finding

The skill description is extremely broad and includes generic domains like work productivity, Google, workspace, CLI, analysis, and implementation support. That can cause the router to invoke this skill for many ordinary requests outside its proven scope, leading to inappropriate tool selection, prompt-surface expansion, and accidental handling of sensitive workflow or workspace tasks under an ill-defined mandate.

Content

No source excerpt is available for this finding.

Vague Triggers

High
Category
Not specified by scanner
Confidence
98% confidence
Finding

The trigger keywords include highly generic terms such as google, workspace, cli, gmail, calendar, drive, contacts, and bug fix without any scope guards. In an agent ecosystem, this makes accidental activation very likely for unrelated or sensitive enterprise requests, increasing the chance of over-collection of context, misrouting, or execution of the wrong workflow against user productivity data.

Content

No source excerpt is available for this finding.

Vague Triggers

High
Category
Not specified by scanner
Confidence
97% confidence
Finding

触发关键词包含非常宽泛且高频的通用词，如“google”“workspace”“cli”“bug fix”，会让该技能在大量与其真实能力边界无关的请求中被错误触发。错误路由会导致用户请求被不恰当地交给此技能处理，增加错误建议、越权作用域扩张或覆盖更合适专业技能的风险。

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The trigger phrases are broad, generic help-seeking prompts that can cause the skill to activate in contexts the user did not clearly intend. In an agent ecosystem, this increases the risk of prompt-routing hijacks or accidental invocation, especially because terms like 'help me' and 'I need a practical workflow' overlap with ordinary requests.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
84% confidence
Finding

The file labels separate English and Chinese documents, while this README itself is specifically the zh-CN variant and does not state that language selection is optional or user-driven. Under the policy, locale-specific behavior should either provide user choice or clearly justify the restriction.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

该文件为 SKILL.zh-CN.md，正文持续以简体中文面向用户提供说明，但未见任何允许用户选择语言或说明仅适用于特定中文区域用户的文字。根据语言/locale 政策，若强制特定语言而无用户 opt-in，属于自然语言策略风险。

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

技能描述中的启用条件覆盖“work-productivity, gog, google, workspace, cli”以及“需要流程、清单、分析或实现支持”等模糊表述，边界过宽且缺乏排除条件。这会让编排器难以区分该技能与其他通用生产力、Google 或 CLI 类技能，造成误调用、提示污染和结果不可靠。

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The default prompt uses a very broad natural-language trigger phrase tied to common productivity terms, which increases the chance of unintended or implicit invocation during ordinary user requests. Because the policy also enables implicit invocation, this can cause the skill to activate outside user intent, expanding the attack surface for prompt injection, unintended tool use, or accidental handling of sensitive workflow content.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The trigger examples are broad, natural-language phrases that can match common user requests without clearly requiring Google Workspace or this skill's specific workflow. That can cause unintended activation and route unrelated tasks into this skill, increasing the chance of incorrect actions, confused delegation, or misuse of any downstream automations tied to the skill.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
71% confidence
Finding

The file declares separate English and Chinese instruction and guide files, but does not state that the user's preferred language will be respected. Because language handling is described in natural language and no opt-in or choice behavior is documented here, this may conflict with a policy requiring language or locale choice rather than implicit forcing.

Content

No source excerpt is available for this finding.