Security audit
Security checks for vulnerabilities and agentic risk
This is a workflow/checklist helper with overly broad trigger wording, but it contains no executable code, persistence, credential handling, or destructive behavior.
Before installing, expect a documentation-oriented helper for skill-vetting workflows. The main caution is that its trigger wording is too generic and could cause accidental activation on ordinary GitHub, security, or bug-fix requests; narrowing triggers would improve usability and reduce routing confusion.
The skill description includes extremely broad activation terms like "security" and "first," which can match many unrelated user requests and cause the skill to activate outside its intended scope. In an agent ecosystem, overbroad routing can lead to inappropriate invocation of this skill, potentially bypassing more suitable or safer specialized skills and degrading trust in workflow selection.
The listed trigger keywords are vague and include common terms such as "security," "first," and "before," making accidental or overly broad activation likely. This is dangerous because an attacker or even normal user phrasing could steer the agent into this workflow unexpectedly, creating prompt-routing confusion and increasing the chance of irrelevant, unsafe, or lower-quality outputs.
The trigger keywords are overly broad and include generic terms like "security", "first", "before", and "github", which can cause the skill to activate in many unrelated conversations. In an agent environment, this can misroute tasks, override more appropriate skills, or inject irrelevant workflow guidance into sensitive contexts, reducing reliability and potentially interfering with safer handling paths.
The trigger sentences are broad, generic phrasings that overlap with normal user requests about security, GitHub, bug fixes, or practical workflows. This can cause the skill to activate in unintended contexts, leading to misrouting, over-application of the skill, or accidental processing of sensitive requests under the wrong workflow.
The trigger phrases are so broad that they can activate on ordinary requests containing generic terms like security, first, github, or bug fix, causing the skill to run outside its intended scope. In an agent environment, overscoped activation can route unrelated user tasks through this workflow, increasing the chance of inappropriate guidance, accidental data exposure to the skill context, or security-analysis actions being applied where they were not requested.
The trigger phrases and keywords are broad enough to cause unintended or overly frequent activation, especially around generic terms like 'security', 'github', 'first', or 'bug fix'. In an agent ecosystem, this can route unrelated user requests into this skill, causing incorrect handling, workflow confusion, or unsafe overreach into security-sensitive tasks without clear user intent.
The example triggers are vague, repetitive, and do not distinguish valid activations from invalid ones, so they reinforce ambiguous routing behavior rather than constraining it. Without specificity, maintainers and automated systems have little guidance for safe invocation boundaries, which increases misfires and makes abuse or accidental overuse easier.
The file is presented only in a Chinese locale version without offering language choice or clarifying fallback behavior. In multilingual agent ecosystems, this can cause operators or downstream systems to misunderstand the skill’s scope, triggers, and safety constraints, which may indirectly increase misconfiguration or misuse risk.
The description says to use the skill when users ask about broad topics like work-productivity, security, or need practical workflows and analysis, but it does not define clear boundaries for when this specific skill is appropriate. That ambiguity increases the chance of accidental invocation in unrelated requests, leading to incorrect task routing and unreliable agent behavior.
The default prompt is broad and uses common workflow terms like help me, practical help, fixing bugs, and security, which increases the chance of accidental or implicit invocation in unrelated user conversations. Because implicit invocation is enabled, this can cause the skill to activate when the user did not intend it, creating prompt-routing confusion and expanding the skill's influence over ordinary tasks.
The file descriptions specify English and Chinese instruction variants, but the README does not state that users may choose their preferred language or locale. Under the policy, fixed language behavior without opt-in can be a natural-language policy concern unless the locale constraint is clearly documented and justified.
This file is labeled as a Chinese README, but key user-facing description text is presented in English. That can amount to a language/locale policy issue because the document effectively forces a language choice that is inconsistent with the stated locale and provides no opt-in or alternative at the point of use.
The file is a zh-CN user document, but all example trigger phrases are written in English. This imposes a specific language for invocation without documenting that users may choose their preferred language or that English is required for a justified reason.
The evidence list includes Chinese-language titles such as "HarmonyOS 开发者社区" and "ONES 研发管理" without any note that multilingual content may appear or that the skill is intended for a Chinese-language context. This can conflict with a language/locale policy when users are not given a choice or expectation about locale-specific material.
