Security audit
Security checks for vulnerabilities and agentic risk
This is a documentation-only workflow helper with overly broad activation wording, but it does not execute code or request sensitive access.
Installers should be aware this skill may activate on ordinary search or bug-fix requests because its trigger language is broad. It appears safe from a security standpoint as a documentation-only helper, but narrowing invocation phrases would make it less disruptive in multi-skill environments.
The description says to use the skill whenever a user asks for broad concepts like 'work-productivity', 'multi', 'search', or 'engine', which creates an ambiguous activation boundary. This makes accidental invocation likely and can misroute user requests to a skill that is only loosely relevant, reducing reliability and potentially exposing downstream workflows to unintended prompt context.
The trigger keywords include extremely generic terms like 'multi', 'search', 'engine', 'integration', and 'supports', which can cause the skill to activate for a wide range of unrelated requests. Over-broad activation increases the chance that this skill intercepts prompts outside its intended scope, leading to incorrect tool routing, prompt hijacking opportunities, or suppression of more appropriate skills.
The description says the skill should be used whenever users mention broad concepts like work-productivity, multi, search, or engine, or whenever they need practical workflow or analysis support for the requirement. This is overly expansive and ambiguous, increasing the chance of accidental invocation on unrelated tasks and weakening control over tool or skill selection.
The trigger keywords include extremely broad generic terms such as "multi", "search", and "engine", which can match many unrelated user requests. This can cause the skill to activate outside its intended scope, potentially overriding more appropriate skills and creating unsafe or misleading workflow routing.
The default prompt contains a very broad invocation phrase tied to generic terms like work productivity, multi, search, and workflow-style assistance. Combined with allow_implicit_invocation: true, this increases the chance the skill is triggered during ordinary user requests, causing unintended routing, prompt injection surface expansion, or accidental execution in contexts the user did not clearly intend.
The trigger sentence starts with a very common phrase ('Help me') and then appends broad requirement text, which can cause the skill to activate for ordinary user requests that were not intended to invoke this workflow. Over-broad activation increases the chance of misrouting user tasks, unintended tool/skill execution, and prompt-surface expansion across unrelated conversations.
The trigger phrases are generic enough to activate on common terms like 'multi', 'search', 'engine', 'integration', and 'bug fix', which can cause the skill to run outside its intended scope. In an agent ecosystem, overly broad activation increases the chance of unintended invocation, workflow hijacking, and misrouting user tasks to this skill when another tool or a safer specialized workflow should handle them.
The invocation guidance lists generic keywords like 'multi', 'search', 'engine', and 'bug fix' without defining scope boundaries, making accidental or excessive invocation likely. This is dangerous because agent selection logic may overmatch on common terms and apply this skill in contexts where its workflow, assumptions, or outputs are not appropriate.
The trigger phrases are broad enough to match ordinary user requests such as asking for workflow help, bug fixes, or search-related assistance, which can cause the skill to activate outside its intended niche. In an agent ecosystem, overbroad activation can route unrelated tasks through this skill, increasing the chance of incorrect handling, unintended instruction injection from skill content, or unsafe delegation patterns.
The trigger section provides only broad keywords and affirmative examples, but no boundaries or negative examples to constrain activation. Without specificity, the skill may match many ordinary requests containing common terms, causing unintended activation and making routing behavior unpredictable in multi-skill environments.
The example trigger phrases repeat the broad requirement language and do not provide contrasting non-trigger examples or boundary conditions. As a result, they fail to constrain matching behavior and reinforce overbroad activation patterns, making accidental or inappropriate routing more likely.
The invocation example uses a generic request pattern ('I need a practical workflow for ...') without strong trigger boundaries, making accidental or adversarial activation more likely. In an agent environment, ambiguous invocation rules can let unrelated prompts match this skill, leading to unreliable routing and possible chaining into unintended actions.
This file is presented as a Chinese README but substantial user-facing content, including the title, requirement description, and trigger phrases, is in English. That imposes a language expectation without telling users whether they may interact fully in Chinese, English, or choose their preferred language.
