Security audit
Security checks for vulnerabilities and agentic risk
This is a documentation-only workflow helper with no executable payload, but its automatic activation wording is broader than necessary.
Installers should be aware that the skill may activate too easily for generic security or GitHub-related requests. Prefer explicit invocation for skill-vetting tasks, and narrow the trigger keywords if maintaining this package.
The skill description is broad enough to match generic terms like 'security' and 'first', which can cause unintended activation for unrelated requests. In an agent routing context, overbroad triggers can misroute sensitive tasks, override more appropriate skills, and expose users to incorrect or incomplete workflows for security-relevant decisions.
The keyword list includes common words such as 'security', 'first', 'before', 'github', and 'bug fix', which are too generic to safely identify this specialized workflow. This makes accidental invocation likely across many unrelated prompts, especially in environments where skill selection is automatic, leading to unreliable routing and potentially unsafe advice in contexts the skill was not designed to handle.
The trigger keywords include very broad generic terms such as "security", "first", "before", and "github", which are likely to appear in many unrelated conversations. This can cause accidental invocation of the skill outside its intended context, leading to inappropriate workflow takeover, user confusion, and possible interference with more suitable skills or safety controls.
The trigger sentence is broad enough to match common user phrasing such as asking for practical help or workflow support, which can cause the skill to activate outside its intended scope. Overbroad activation increases the chance that a security- or vetting-branded skill intercepts unrelated conversations and steers behavior, creating misrouting and trust-boundary problems rather than direct code execution risk.
The trigger remains ambiguous because it uses vague language and truncated phrasing that does not clearly delimit when the skill should or should not run. In practice this can cause accidental invocation on unrelated requests, exposing users to incorrect workflow guidance and making routing behavior easier to manipulate through prompt wording.
The trigger phrases are broad and include common terms like 'security', 'first', 'github', and generic help wording, which can cause the skill to activate for unrelated user requests. In a security-oriented workflow helper, unintended invocation is risky because users may receive vetting or workflow guidance when they did not explicitly ask for this skill, increasing the chance of inappropriate actions, confusion, or over-collection of context.
The trigger phrases are broad enough to match generic terms like "security," "first," "github," and "bug fix," which can cause the skill to activate in contexts far outside its intended scope. In an agentic system, this increases the risk of incorrect routing, accidental invocation, and overexposure of a workflow that may influence security-related decisions without sufficient user intent.
The example trigger sentences are malformed and do not clearly demonstrate the intended invocation boundary, so they fail to constrain usage and may reinforce overly broad matching behavior. Poor examples increase the chance that developers or routing systems interpret vague fragments as valid triggers, compounding accidental activation risk.
The skill description defines usage conditions too broadly, covering general requests for workflow, checklist, analysis, implementation support, and several common keywords without clear boundaries. In an agent-routing environment, this ambiguity increases the chance that the skill is selected for unrelated tasks, producing misleading outputs or bypassing more appropriate specialized skills.
The default prompt uses broad, everyday phrasing such as general requests for help, workflow, checklist, analysis, or implementation support, making the skill eligible for activation in many ambiguous situations. In combination with implicit invocation, this increases the risk of over-triggering, prompt hijacking of routing behavior, and accidental use in contexts where the user did not intend to engage this skill.
The skill enables implicit invocation while providing no narrowly defined activation constraints, which can cause the agent to invoke this skill in loosely related contexts without explicit user intent. Because the skill is framed around security, vetting, and workflow help, overbroad automatic routing could expose sensitive user context to the skill or cause unintended security-relevant actions and recommendations.
The file descriptions state that the skill instructions and guides exist in English and Chinese variants, but do not indicate how the user's preferred language is selected or whether the user can choose. This can conflict with a language/locale policy if the skill defaults to a language without explicit user opt-in.
The file is a Chinese README, but key user-facing content is presented in English, including the main requirement description. This can amount to an implicit language constraint or inconsistent locale behavior without giving users an explicit language option or justification.
