Security audit
Security checks for vulnerabilities and agentic risk
This is a documentation-only workflow helper with no executable code or persistence, but its activation wording is overly broad and users should enable it deliberately.
Install only if you want a general workflow helper for Skill Vetter-style planning and reliability tasks. Because implicit invocation is enabled and the trigger words are broad, consider disabling implicit use or invoking the skill explicitly by name so it does not appear in unrelated security, GitHub, or bug-fix conversations.
The manifest description includes very broad activation terms such as 'security', 'first', and generic requests for workflows, analysis, or implementation support. This can cause the skill to activate on many unrelated user requests, leading to inappropriate invocation, context hijacking, or accidental routing of sensitive tasks into a skill not specifically requested by the user.
The trigger list uses ambiguous keywords like 'security', 'first', 'before', 'github', and 'bug fix' without constraints, making accidental matches highly likely. In an agent ecosystem, this increases the chance of over-broad interception of user intent, misrouting tasks, and unauthorized influence over conversations that only loosely mention these common terms.
The trigger keyword list includes very broad, high-frequency terms such as 'security', 'first', 'before', and 'github', which can cause the skill to activate in many unrelated conversations. Over-broad activation increases the chance that this skill intercepts requests outside its intended scope, potentially overriding more appropriate skills or influencing security-sensitive workflows unexpectedly.
The trigger phrases include broad, common terms such as "security," "first," and "github," which can cause the skill to activate for many unrelated user requests. In an agent environment, overly broad routing increases the chance that this skill intercepts prompts outside its intended scope, leading to incorrect workflow selection, confusing outputs, or accidental handling of sensitive security-related tasks without proper specificity.
The trigger examples are broad, generic, and include common terms like 'security', 'first', and 'github', which can cause the skill to activate in contexts far beyond its intended scope. In an agent system, overbroad activation can route unrelated or sensitive tasks into this workflow, increasing the chance of incorrect handling, prompt collisions, or unsafe automation decisions.
The trigger phrases are broad and include generic terms like 'security', 'first', 'before', and 'github', which can cause the skill to activate in unrelated contexts. In an agent ecosystem, unintended invocation can route user requests into the wrong workflow, causing incorrect guidance, confusion, or unsafe handling of tasks that the user did not intend this skill to perform.
The skill description defines activation conditions broadly ('work-productivity, skill-vetter, vetter, security, first' or any request for workflow/checklist/analysis support), without clear limits on when the skill should not apply. This ambiguity can lead to accidental invocation in unrelated contexts, reducing routing precision and causing the skill to influence tasks beyond its intended domain.
The default prompt and invocation phrase are broad, generic, and overlap with common user requests about productivity, security, workflows, checklists, and implementation help. This increases the chance of accidental routing or over-invocation, causing the skill to activate in contexts the user did not specifically intend, which is especially risky for a security-adjacent helper that may influence analysis or workflow decisions.
Enabling implicit invocation without defined trigger constraints or exclusions allows the skill to be auto-selected during ordinary conversations that mention broad terms like security, checklist, analysis, or workflow. In this skill's context, that can create unintended participation in sensitive or security-related tasks, increasing the risk of user confusion, prompt-scope creep, and unauthorized influence over agent behavior.
This file is labeled as a Chinese README, but substantial operational content such as the requirement description, workflow description, and trigger phrases remains in English. That creates an implicit language policy inconsistency without giving users a language/locale choice or explaining why English is required.
