Security audit
Security checks for vulnerabilities and agentic risk
This is a documentation-only workflow helper with overly broad activation wording, but no evidence of hidden execution, data access, persistence, or destructive behavior.
Before installing, be aware that this skill may activate too broadly for ordinary writing, editing, or bug-fix prompts. Prefer explicit invocation by skill name when you want it, and review outputs normally before applying any suggested code or workflow changes.
The manifest description uses expansive phrases such as 'use when a user asks for' followed by very common terms, which can cause the orchestrator to match this skill against a wide range of unrelated prompts. In practice, this broad invocation surface increases the chance of misrouting and accidental application of the skill in contexts where it may override more appropriate or safer skills.
The trigger keywords are so broad and generic that the skill can activate for many ordinary writing, editing, or productivity requests that are unrelated to this specific workflow. This creates a routing/invocation vulnerability where the agent may select and apply the wrong skill, causing unintended behavior, policy drift, or unsafe transformations in contexts the author did not intend.
The trigger sentence is broad and matches common help-seeking language, which can cause the skill to activate outside its intended scope. In an agent ecosystem, unintended invocation can route unrelated user requests through this skill, producing incorrect behavior, reducing reliability, and potentially bypassing safer or more appropriate skills.
The README describes when to use the skill with vague keywords and generic task language, but does not define boundaries for when it should not activate. This ambiguity increases the chance of accidental invocation and prompt-collision with other skills, reducing predictability and potentially causing an agent to take unintended actions.
The trigger examples are broad, natural-language phrases that could match many unrelated user requests, causing the skill to activate outside its intended scope. In an agent environment, this can route tasks to the wrong workflow, produce irrelevant edits or analysis, and create unsafe or confusing automation behavior.
The trigger phrases are broad, generic, and likely to match ordinary user requests unrelated to this specific skill. That can cause unintended invocation of the skill in contexts where it was not requested, increasing the risk of prompt-routing mistakes, irrelevant behavior, or misuse of workflow guidance in unrelated tasks.
The example triggers rely on generic lead-ins like 'Help me' and 'I need a practical workflow' and do not sufficiently constrain subject matter, making them poor discriminators for skill selection. These examples can train or bias a selector toward invoking the skill for broadly phrased user requests, increasing false activations and workflow confusion.
The skill description says it should be used when users ask for broad terms such as work-productivity, humanizer, remove, signs, or generated, or when they need general workflow/checklist/analysis support for the requirement. This activation guidance is ambiguous and insufficiently bounded, making it easy for the system to invoke the skill for loosely related prompts and produce unintended behavior.
The trigger keywords are overly broad and include common words like 'remove', 'generated', 'writing', 'text', and 'reviewing', which can match many unrelated user requests. This increases the chance of unintended skill activation, causing the agent to apply the wrong workflow or expose users to irrelevant automation in contexts where it was not requested.
The manifest description and default prompt use very broad terms such as 'help me' and generic workflow/support language, which can match many ordinary user requests unrelated to this specific skill. This increases the chance of inappropriate routing or silent invocation, causing the agent to apply this skill in contexts the user did not intend and potentially exposing them to unintended transformations or guidance.
Enabling implicit invocation without tightly scoped trigger conditions allows the platform to auto-select this skill for loosely related requests. In combination with the broad manifest wording, this raises the risk of overbroad activation, unexpected behavior, and user confusion about why the skill was invoked.
This trigger is ambiguous because it does not clearly define the activation boundary and relies on vague wording that could match many unrelated requests. Such ambiguity increases the chance of accidental routing and misapplication of the skill, especially in automated dispatch systems that depend on trigger specificity.
This Chinese-language README contains substantial English instruction-like content describing the requirement and workflow behavior. Because no language preference or opt-in is offered, the file may effectively impose a language/locale choice on users despite being presented as a localized document.
