Security audit

Work Productivity Gog Google Workflow Helper

Security checks for vulnerabilities and agentic risk

Overview

This skill is a low-risk workflow helper with overly broad activation wording but no code, credentials, persistence, or hidden actions.

Install only if you want a general Gog or Google Workspace workflow-planning helper. Be aware that its broad implicit triggers may activate for ordinary Google, Gmail, Drive, Calendar, CLI, or bug-fix requests, so users or maintainers should narrow the activation wording before relying on automatic routing.

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

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 (13)

Vague Triggers

High

Category: Not specified by scanner
Confidence: 96% confidence
Finding: The skill description is extremely broad and includes generic terms like 'google', 'workspace', 'cli', 'analysis', and 'implementation support', which can cause the skill to activate for many unrelated requests. Unintended invocation can route users into an irrelevant workflow, increasing the chance of confused-deputy behavior, bad task scoping, and inappropriate handling of sensitive productivity data.
Content: No source excerpt is available for this finding.

Vague Triggers

High

Category: Not specified by scanner
Confidence: 98% confidence
Finding: The trigger list contains ambiguous everyday keywords such as 'google', 'workspace', 'cli', 'gmail', 'calendar', and 'drive' without scope constraints or disambiguation. Because these terms appear in many benign and unrelated contexts, the skill may be invoked unnecessarily and gain access to conversations involving potentially sensitive business communications or files when a more appropriate skill should handle them.
Content: No source excerpt is available for this finding.

Vague Triggers

High

Category: Not specified by scanner
Confidence: 95% confidence
Finding: 技能描述把启用条件写成“当用户提出 work-productivity, gog, google, workspace, cli，或需要围绕该需求获得实用流程、产物、检查清单、分析或实现支持时使用”，范围明显过宽且边界模糊。这样的描述容易覆盖大量常规生产力、Google、CLI 相关请求，使该技能在不相关对话中被优先选中，降低路由准确性并放大后续输出偏差。
Content: No source excerpt is available for this finding.

Vague Triggers

High

Category: Not specified by scanner
Confidence: 98% confidence
Finding: 触发关键词包含 google、workspace、cli、bug fix 等高频且语义宽泛的词，会让技能在大量普通办公或开发场景中被误触发。误触发会把模型路由到与用户真实意图不完全匹配的技能，增加越权介入、错误建议或提示注入面扩大等风险。
Content: No source excerpt is available for this finding.

Vague Triggers

High

Category: Not specified by scanner
Confidence: 97% confidence
Finding: The trigger sentences use broad, ordinary phrasing such as 'I need a practical workflow' that can match many unrelated requests. In an agent-routing system, this can cause accidental invocation of the skill outside its intended scope, leading to misrouting, inappropriate tool use, or unexpected handling of user data in Google/workspace-related contexts.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The trigger phrases are broad, generic, and include common productivity terms and templated language that could cause the skill to activate in contexts far beyond its intended scope. Over-broad activation increases the chance of unintended invocation, misrouting user requests, and exposing users to actions or guidance they did not mean to request, which is a real security and safety concern for agent skills.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The trigger phrases are broad, generic, and overlap with common productivity requests such as Google Workspace, Gmail, Calendar, Drive, CLI, and bug fixing. This can cause the skill to activate unintentionally in unrelated conversations, expanding its operational scope and increasing the chance that users are routed into the wrong workflow or that adversarial prompts exploit the skill’s wide matching surface.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 90% confidence
Finding: The example trigger sentences repeat vague demand-language and encourage activation from broad problem statements rather than precise user intent. Without counterexamples or non-trigger cases, integrators are more likely to over-match, leading to accidental invocation and reduced trust in routing accuracy.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 90% confidence
Finding: 示例触发句使用了非常通用的请求模板，如 Help me...、I need a practical workflow...，但没有提供反例来说明什么情况下不该触发。缺少负面示例会让实现方或路由器更难学习边界，进一步提高误匹配概率。
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The default prompt contains a long, generic trigger phrase tied to common concepts like work productivity, workflow help, and practical support, which can cause the skill to be invoked in unrelated conversations. Because implicit invocation is enabled, this increases the chance of unintended routing or prompt injection surface expansion whenever normal user language overlaps with the trigger text.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 80% confidence
Finding: One evidence item includes the phrase "Seed de conteudo," indicating mixed or non-English language content without any user choice or locale explanation elsewhere in the file. Because the skill materials do not offer language selection or justify locale-specific behavior, this may conflict with language/locale policy expectations.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 91% confidence
Finding: The trigger scope is ambiguous because the activation criteria blend broad category keywords with long requirement text that is not operationally precise. This makes the skill easier to invoke unintentionally or by prompt injection-style phrasing, increasing the chance that unrelated user requests are routed into a workflow that assumes Google/work-productivity relevance.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

Category: Not specified by scanner
Confidence: 81% confidence
Finding: This file is labeled as a Chinese README, but substantial user-facing instructional content and trigger phrases remain in English. That can effectively force a language preference on users without opt-in or explanation, which may conflict with language/locale policy expectations.
Content: No source excerpt is available for this finding.