Security audit
Security checks for vulnerabilities and agentic risk
This is a documentation-only workflow helper; its activation wording is overly broad, but I found no code execution, persistence, credential use, data collection, or destructive behavior.
This skill appears safe to install from a security perspective, but users should be aware it may activate too often because its keywords are generic and implicit invocation is enabled. Prefer explicit invocation or narrowed triggers if predictable routing matters.
The trigger phrases are broad enough to match many ordinary user requests, which can cause the skill to activate outside its intended scope. In an agent ecosystem, overbroad activation can hijack unrelated conversations, inject workflow behavior unexpectedly, and increase the chance that a lower-trust skill influences sensitive tasks.
The trigger list is excessively broad and includes generic words like 'security', 'every', 'must', 'before', and 'bug fix' that commonly appear in unrelated requests. This can cause the skill to activate in many contexts outside its intended scope, leading to prompt hijacking of routing, user confusion, or inappropriate application of the skill in security-sensitive workflows.
The description says the skill should be used when a user asks for broad categories like work-productivity, skillscan, security, gate, or every, or when they need general workflow or analysis support for the requirement. This creates an ambiguous activation boundary that can match many unrelated requests, increasing the chance of accidental invocation and misrouting of sensitive or irrelevant user tasks into this skill.
The trigger list includes very generic terms such as "security", "gate", "every", "must", "pass", and "before", which are common in ordinary conversations and likely to activate the skill unintentionally. Over-broad activation can cause the wrong skill to run, leading to confused task routing, exposure of unrelated context to the skill, and reduced reliability of downstream security workflows.
The manifest description and default prompt are extremely broad, with trigger terms like "work-productivity," "security," "gate," and "every," which can cause the skill to activate in many unrelated contexts. This increases the chance of unintended routing, prompt-surface expansion, and privilege confusion where users or orchestrators invoke the skill outside its intended scope.
Enabling implicit invocation without narrow gating allows the platform to auto-select this skill based on weak or ambiguous matches. Combined with the broad manifest wording, this can lead to accidental invocation in unrelated conversations, exposing users to unintended behavior and increasing the attack surface for prompt injection or workflow interference.
The trigger content includes extremely broad terms like "security," "gate," "every," and "must pass before activate," which can match many ordinary user requests unrelated to this skill’s intended scope. In an agent environment, overbroad activation can cause unintended routing, privilege expansion of the wrong workflow, or interference with safer/more appropriate skills, especially because this skill is positioned as a security-oriented helper.
The trigger phrases are broad, generic, and include common workflow/security terms like "security," "gate," "every," "must," and "bug fix," which can cause the skill to activate in contexts far outside its intended scope. In an agent ecosystem, overly broad activation boundaries can lead to unintended routing, shadowing of more appropriate skills, and unsafe application of this skill’s instructions to unrelated security-sensitive tasks.
This zh-CN README presents the usage section and activation examples primarily in English, which effectively imposes a language preference in a locale-specific document. There is no indication that users may choose Chinese or English trigger phrasing, nor any justification for requiring English.
The keyword list includes generic terms such as 'security', 'every', 'must', 'pass', and 'bug fix', which are common across unrelated requests. This creates ambiguous routing conditions that may spuriously invoke the skill, reducing predictability and potentially exposing users to unintended instructions during unrelated workflows.
The invocation description is broad and ambiguous, saying to use the skill whenever a user asks for common concepts like workflow, checklist, analysis, or implementation support related to several generic terms. This weak scoping increases the chance the orchestrator selects this skill for unrelated prompts, which can override more appropriate skills or inject unintended behavior into normal user interactions.
The sample trigger sentence is generic enough that normal requests for a "practical workflow" could invoke this skill even when the user is asking for unrelated productivity help. That increases the chance of misrouting, confusing outputs, and accidental execution of a security-themed workflow in contexts where it does not belong.
