Security audit
Security checks for vulnerabilities and agentic risk
This is a documentation-only workflow helper with no code execution or persistence, though its auto-invocation triggers are too broad and users should scope when it runs.
Install only if you want a general SkillScan-style workflow helper, and consider tightening or disabling implicit invocation so ordinary security or productivity requests do not route to this skill unintentionally.
The skill description says to use the skill when a user asks for very broad terms like "work-productivity," "security," or generic workflow help, which can cause the skill to activate outside its intended scope. Over-broad routing increases the chance of unintended instruction injection, wrong-tool execution, or security-relevant tasks being handled by an inapplicable skill.
The keyword list includes highly generic words such as "security," "gate," "every," "must," "pass," and "before," which are common in ordinary conversation and likely to match unrelated user prompts. This can lead to accidental activation and context hijacking, especially in systems that auto-select skills based on keyword overlap.
The trigger keyword list includes very broad terms such as "security", "every", "must", "pass", and "before", which are common in unrelated conversations. This can cause accidental activation of the skill outside its intended scope, leading the agent to apply the wrong workflow or override more appropriate skills during security-sensitive tasks.
Enabling implicit invocation without tight activation constraints allows the platform to auto-select this skill based on vague language matches rather than explicit user consent. Because the skill is framed around broad security and workflow assistance, this increases the risk of unsolicited activation, instruction interference, and accidental exposure of user tasks or context to the skill in unrelated conversations.
The trigger guidance includes the everyday term "every," which is so broad that it can cause the skill to activate for many unrelated prompts. In an agent ecosystem, overbroad triggers can hijack routing, cause unintended execution, and suppress more appropriate skills, creating reliability and security issues through prompt capture rather than code execution.
The keyword list includes highly generic terms such as 'security,' 'every,' 'must,' 'pass,' 'before,' and 'activate,' which are likely to appear in many unrelated prompts. This creates a substantial risk of accidental invocation or shadowing of more appropriate skills, which is especially concerning for a security-themed workflow that may be trusted to influence gating or review decisions.
The trigger examples are phrased as generic help requests rather than narrow, distinctive invocations, so the skill may activate in response to ordinary user language unrelated to this specific workflow. In an agent ecosystem, over-broad activation can route sensitive or unrelated tasks into this skill unexpectedly, causing incorrect handling, policy bypass through misrouting, or unsafe automation decisions.
The trigger keywords and example phrases are broad enough to match common terms like "security," "workflow," "every," or "before," which can cause unintended activation in unrelated conversations. In an agent skill context, overbroad invocation increases the chance that the wrong skill runs on sensitive requests, leading to incorrect automation, policy bypass by misrouting, or unnecessary exposure of user context to this skill.
The example trigger sentences are malformed, vague, and resemble ordinary prose rather than clear invocation patterns, making it harder for users and routers to distinguish intentional use from incidental text. Ambiguous examples reinforce over-triggering behavior and can cause the skill to be selected during unrelated conversations.
The skill description says to use the skill when users mention broad categories like work-productivity, skillscan, security, or need practical support, but it does not clearly define boundaries for when it should not activate. In an agent routing context, ambiguous activation criteria can misroute requests, especially when the topic overlaps with other security or productivity skills.
The example trigger phrases are all positive examples and are themselves broad, offering no negative examples or disambiguation guidance. This increases the chance that downstream routing systems generalize too aggressively and invoke the skill for loosely related requests.
The default prompt uses a very broad natural-language trigger phrase that overlaps with common user requests such as productivity, workflow, checklist, or analysis help. This can cause the skill to be invoked in contexts the user did not specifically intend, increasing the chance of prompt-scope confusion, unintended tool influence, or accidental routing of sensitive tasks through this skill.
The invocation guidance mixes specific keywords with broad phrasing and sentence templates that could match routine user requests unrelated to this skill. That increases the chance of accidental or excessive activation, which is particularly risky for a security-oriented helper because it may insert itself into workflows where its assumptions do not fit and where other controls should apply.
This file is presented as a Chinese README, but the operational usage content and trigger phrases are written in English only. That creates an implicit language constraint without user opt-in and may violate language/locale expectations for a localized skill document.
