Security audit
Security checks for vulnerabilities and agentic risk
This skill is a documentation/workflow helper with no code execution, credential handling, persistence, or hidden data movement, though its activation wording is overly broad.
Installers should expect a lightweight workflow helper. The main practical caveat is that implicit invocation plus generic trigger words may cause this skill to activate for unrelated search or productivity requests, so narrowing the trigger terms would improve predictability.
The trigger keyword list includes extremely generic terms like "multi", "search", "engine", "integration", and "supports", which can cause the skill to activate for many unrelated requests. Over-broad activation increases the chance that the wrong skill handles user input, creating prompt-routing confusion and potentially exposing users to inappropriate instructions or unintended workflow execution.
The trigger keywords include extremely broad generic terms such as "multi", "search", and "engine", which can match many unrelated user requests and cause accidental skill activation. In an agent environment, over-broad routing can divert tasks into an unintended workflow, producing irrelevant actions, confused outputs, or unsafe automation in the wrong context.
The trigger sentence is overly broad and can cause the skill to activate on common, everyday phrasing rather than a clear request for this specific workflow. In an agent ecosystem, ambiguous activation can route unrelated user tasks into this skill, leading to incorrect tool usage, unintended workflow execution, or prompt-scope confusion that degrades safety and reliability.
The trigger phrases are broad, generic, and partially malformed, making it easy for unrelated user requests containing words like 'multi', 'search', or 'engine' to activate the skill unintentionally. In an agent environment, ambiguous routing can cause the wrong workflow to run, producing incorrect actions, unnecessary tool use, or unsafe handling of user requests outside the skill's intended scope.
The documented trigger keywords and example phrases are extremely generic terms like "multi", "search", "engine", and broad help-seeking language, which can cause the skill to activate for unrelated user requests. Overbroad activation increases the chance of unintended invocation, context hijacking, and incorrect workflow execution, especially in agent ecosystems where routing decisions may be automated.
The description says to use the skill when a user asks for broad concepts like work-productivity, multi, search, or engine, plus generic needs like a workflow or checklist. This vague invocation criteria can make the skill match unrelated tasks, weakening routing integrity and reducing predictability of agent behavior.
The example trigger sentences repeat broad language and do not clarify what should not match, so they reinforce ambiguous routing. Without clear boundaries or negative examples, downstream systems or authors may overgeneralize invocation patterns and attach this skill to unrelated user requests.
The description says to use the skill when a user asks for broad terms like work-productivity, multi, search, or engine, without clearly constraining the request to this specific workflow domain. This weak boundary makes accidental invocation likely and increases the chance that the agent applies this skill in unrelated contexts where its assumptions and instructions do not fit.
The example trigger phrases demonstrate invocation patterns but do not show any limiting conditions or disambiguation guidance, so they reinforce broad matching behavior rather than reducing false activations. This can train maintainers or routing logic to treat vague requests as valid matches, compounding the over-triggering problem created elsewhere in the skill.
The default prompt is phrased so broadly that common user language around work, search, workflows, or implementation help could unintentionally trigger this skill. That creates prompt-scope confusion and raises the chance of unauthorized or unexpected skill activation, which can steer conversations or actions without the user clearly intending to invoke this specific helper.
Enabling implicit invocation without tightly scoped activation criteria allows the agent platform to invoke this skill in response to ambiguous requests. In a broadly described productivity/search context, that increases the risk of accidental routing, unintended behavior, and abuse through crafted prompts that cause the wrong skill to activate.
This trigger is ambiguous about when the skill should be invoked, so an orchestrator or downstream agent may match on vague language and select the skill outside its intended scope. That increases the chance of misfires, irrelevant execution, and accidental handling of tasks the skill is not designed to perform, which is especially risky in automated multi-skill workflows.
This file is labeled as a Chinese README, but key user-facing content such as the title and requirement description is presented primarily in English. That can amount to an implicit language policy choice without user opt-in or justification.
