Security audit
Security checks for vulnerabilities and agentic risk
This is a documentation-only workflow helper with overly broad trigger wording but no hidden code, credential access, persistence, or destructive behavior.
Installers should be aware that this skill may trigger too broadly because of generic keywords and implicit invocation. It is otherwise a low-risk documentation/workflow helper; consider narrowing triggers before publishing broadly so it activates only for explicit multi-search-engine workflow tasks.
The trigger list includes extremely generic terms such as "multi," "search," "engine," and "bug fix," which are common in many unrelated conversations. In an agentic environment, this can cause unintended skill activation and route user requests into a workflow the user did not intend, creating prompt-routing confusion and increasing the chance of irrelevant or unsafe automation being applied.
The trigger keyword list includes overly broad generic terms such as “multi”, “search”, and “engine”, which can cause the skill to activate for many unrelated user requests. This creates routing ambiguity and may unexpectedly invoke the skill in contexts it was not designed for, reducing reliability and potentially exposing users to inappropriate workflow guidance.
The trigger sentence begins with a broad everyday request pattern ('Help me ...'), which can cause the skill to activate for loosely related or unintended user prompts. In an agent environment, overly broad routing increases the chance of misfires, context hijacking, or the wrong skill handling sensitive tasks, reducing predictability and safety.
The trigger phrases and keywords are broad enough that the skill may activate on generic terms like "multi," "search," or "engine," causing unintended invocation outside the intended workflow. In an agent ecosystem, ambiguous activation can route unrelated user requests into this skill, leading to incorrect task handling, reduced reliability, and possible unsafe cross-context behavior if the skill is used where it does not belong.
The trigger phrases are broad and generic enough to match ordinary user requests involving productivity, search, integration, or bug fixing, which can cause the skill to activate outside its intended scope. In an agent environment, overbroad activation can route unrelated tasks through this skill, increasing the chance of unintended behavior, prompt interference, or policy bypass through accidental invocation.
The description says to use the skill when a user asks for broad concepts like work-productivity, multi, search, or engine, which does not clearly define when activation is appropriate. Ambiguous activation conditions can lead to misrouting and over-selection of this skill, especially in systems that automatically choose skills based on descriptions.
The example trigger sentences are effectively templates that match very broad language and do not define boundaries for legitimate invocation. This makes the examples poor discriminators for routing and can reinforce accidental activation whenever related generic wording appears in user prompts.
The skill description defines activation conditions too broadly, including generic productivity/search-related requests and vague requests for any workflow, checklist, analysis, or implementation support around the requirement. Such unclear boundaries can over-match normal conversations, causing accidental activation and misrouting to this skill instead of a more appropriate one.
The default prompt uses a very broad, natural-language trigger phrase ('Use $work-productivity-multi-search-workflow-helper to help me ...') tied to common productivity and search-related requests, while the policy also enables implicit invocation. This combination increases the chance the skill is activated in situations the user did not explicitly intend, which can cause unintended tool use, prompt injection exposure through retrieved content, or over-broad delegation to the skill.
Although the sentence includes requirement-related text, it still starts with a generic request form ('I need a practical workflow for ...'), making it ambiguous and prone to matching ordinary user language outside the intended scope. This can broaden skill selection beyond relevant cases, leading to unintended invocation and potentially unsafe or confusing task execution.
The file explicitly declares only English and Chinese instruction and guide files, which may indicate a fixed language/locale offering rather than user language choice. Because no opt-in or broader language selection is described here, this can conflict with a policy against forcing a specific language without user choice.
In a Chinese-language README, the documented trigger phrases are presented only in English (for example, "Help me" and "I need"). This can effectively force a specific language for invocation without stating that users may choose their preferred language or that English is required for a justified reason.
