Security audit
Security checks for vulnerabilities and agentic risk
This is a documentation-only workflow helper with broad activation wording, but it does not request sensitive access, persistence, code execution, or data exfiltration.
Installing this skill should be low risk, but users should be aware it may activate too easily on generic security, GitHub, or bug-fix requests. Prefer explicit invocation for Skill Vetter-style workflow help, and consider narrowing trigger keywords before publishing broadly.
The skill description includes highly generic activation terms such as 'security' and 'first' and frames the skill for broad practical help, which makes accidental routing likely. Over-broad invocation can cause the wrong skill to activate, leading to irrelevant or lower-safety guidance being applied in contexts the author did not intend.
The trigger keyword list contains common words like 'security', 'first', 'before', and 'github' without contextual constraints. In an agent ecosystem, these terms are likely to match unrelated user requests, causing unintended skill invocation and potentially overriding more appropriate, task-specific handling.
The trigger keywords are overly broad and include generic terms like "security", "first", "before", and "github", which can match many unrelated user requests. This can cause the skill to activate in unintended contexts, leading to workflow hijacking, irrelevant guidance, or displacement of a more appropriate skill during security-sensitive interactions.
The default prompt includes very broad trigger language such as 'help me' and generic productivity/security workflow terms, which can overlap with ordinary user requests and cause unintended invocation. Because implicit invocation is enabled, this increases the chance the skill is pulled into conversations without clear user intent, potentially exposing users to unexpected behavior or prompt influence.
The trigger sentences are written in broad, natural language that overlaps with ordinary user requests, making accidental or overly eager invocation likely. In a security- and vetting-oriented skill, unintended activation can route users into the wrong workflow, create misleading trust signals, or cause the agent to act on weak relevance matches.
The trigger phrases are broad enough to match common words like 'security', 'first', 'github', and generic requests for workflows or bug fixes, which can cause unintended skill invocation. In an agent ecosystem, this increases the chance the wrong skill activates and influences sensitive decisions such as security vetting or installation guidance without clear user intent.
The trigger phrases are overly broad and include generic terms such as 'security', 'first', 'before', 'installing', and 'github', which are common in ordinary user conversations. This can cause the skill to activate unintentionally in unrelated contexts, potentially injecting workflow behavior or guidance when the user did not request this specific skill, increasing the risk of misrouting and confusing security-sensitive interactions.
The example trigger sentences reinforce vague activation behavior by showing malformed, catch-all prompts rather than clearly scoped examples. This trains integrators or routing logic toward permissive matching, increasing the chance of accidental or inappropriate activation.
The description defines activation conditions too broadly, saying the skill should be used whenever users ask for general areas like work productivity, security, or practical support artifacts. In a multi-skill environment, this ambiguity increases the chance of inappropriate routing and can expose users to incorrect or incomplete security guidance when a more specialized skill should handle the request.
The keyword list includes very generic terms such as 'security', 'first', and 'github', which are likely to appear in many unrelated conversations. This increases the chance of mis-triggering the skill and is more concerning here because the skill presents itself as a workflow helper for vetting and security decisions, where accidental invocation may bias agent behavior or displace better-matched skills.
The file lists English and Chinese variants of the documentation but does not state how language is selected or that the user can choose their preferred locale. This can create ambiguity around language behavior and may conflict with a language-choice policy if the skill defaults to a specific language without opt-in.
This zh-CN README presents core user-facing content in English despite being the Chinese-language variant, and it does not state that the user can choose their preferred language. That may violate language/locale expectations by implicitly forcing English content in a locale-specific document.
