Security audit
Security checks for vulnerabilities and agentic risk
This is a documentation-only workflow helper with overly broad triggers, but it does not contain code, persistence, credential access, or hidden execution behavior.
Install only if you want a general workflow helper for self-improving or proactive agent productivity tasks. Be aware that its trigger keywords are broad, so users or platform maintainers should prefer explicit invocation by skill name or narrow the trigger rules to avoid accidental activation on ordinary requests.
The description says to use the skill when a user asks for broad categories like work-productivity, self, improving, or proactive, which creates an ambiguous activation rule. Because these terms are common and underspecified, the skill may be invoked for many unrelated prompts, increasing the chance of misrouting and unintended influence over agent behavior.
The trigger list includes extremely broad terms such as "self," "improving," "proactive," and related generic concepts that appear in many ordinary user requests. This can cause the skill to activate outside its intended scope, leading to inappropriate routing, instruction shadowing, or accidental application of workflow behavior to unrelated tasks.
触发关键词包含极其常见且语义宽泛的词，如“self”“improving”“proactive”“learning”“bug fix”，会与大量普通对话重叠，导致技能在并非目标场景下被误触发。误触发会把无关请求路由到该技能，放大不恰当自动化、错误建议或越权工作流介入的风险，尤其该技能定位为“proactive/self-improving”时更容易在边界不清下主动扩展操作。
The trigger phrases are broad, generic, and include common wording like 'help me' and 'I need a practical workflow,' which can cause the skill to activate for unrelated requests. In an agent ecosystem, over-broad invocation can route sensitive or off-topic tasks into this skill unexpectedly, increasing the chance of inappropriate automation, confusing behavior, or bypass of more suitable controls.
The usage section relies on ambiguous natural-language examples instead of tightly scoped activation rules, so other skills or ordinary prompts could match unintentionally. Because this skill targets proactive and self-improving workflows, unintended activation is somewhat more dangerous: it may steer an agent into autonomous planning, workflow changes, or advisory behavior the user did not explicitly request.
The trigger sentence is broad natural language that overlaps with ordinary user requests, making accidental or unintended activation plausible. In an agent ecosystem, ambiguous invocation can cause the wrong skill to run, leading to unexpected actions, misleading outputs, or bypass of the user's intended workflow boundaries.
The keyword list includes extremely broad terms like 'self', 'improving', and 'proactive', which are common in normal conversation and not unique to this skill. This makes accidental or unintended matching likely, especially in automated skill-selection pipelines, and is more dangerous here because the skill is positioned as a workflow helper that may shape execution rather than just provide passive reference content.
The trigger phrases are generic request patterns such as 'Help me' and 'I need a practical workflow' combined with broad problem wording, which can cause the skill to activate for ordinary conversation unrelated to the intended narrow use case. In an agent ecosystem, overbroad activation can route unrelated user requests into this skill, leading to inappropriate behavior, prompt-scope confusion, and increased exposure to downstream unsafe actions if the skill performs proactive or self-improving workflows.
The example trigger phrases are template-like and generic, and they repeat broad language rather than demonstrating precise boundaries for valid activation. This reinforces overbroad matching behavior and makes it harder for integrators or routing systems to distinguish legitimate use cases from normal conversation.
技能描述将适用范围定义为一大类模糊需求，包括“work-productivity”“self”“proactive”以及“practical workflow, artifact, checklist, analysis, or implementation support”，边界非常宽，缺少明确排除条件。这样的描述会使调度器或使用者难以判断何时应调用该技能，增加误路由、过度介入和把普通请求升级为带有自我改进/主动行为工作流的概率。
The manifest allows implicit invocation while describing the skill in very broad, generic productivity/help terms. This increases the chance the platform will auto-select the skill in situations the user did not explicitly request, which can lead to unexpected prompt injection exposure, over-broad assistance, or unintended execution paths in unrelated workflows.
This file is presented as a Chinese README, but key descriptive and trigger content is written in English, which may effectively impose a language expectation on users without opt-in or explanation. That can conflict with a language/locale policy requiring user choice or clear justification for language constraints.
The file includes multilingual natural-language content, including Chinese titles in the evidence list, but does not state whether the skill supports multiple languages or how language selection is handled. This can create an implicit locale expectation without explicit user opt-in or documented justification.
