Security audit
Security checks for vulnerabilities and agentic risk
This is a documentation-only workflow helper with broad activation wording, but it does not contain hidden execution, credential access, persistence, or destructive behavior.
Before installing, be aware that this skill may activate too broadly on generic web-search or workflow requests. It appears safe as a documentation helper, but users or maintainers should narrow the trigger keywords and clarify language behavior to reduce accidental invocation.
The skill description uses very broad activation terms like 'work-productivity', 'web', 'search', 'via', and 'needs a practical workflow' without meaningful scope limits. This can cause the skill to activate on many unrelated user requests, increasing the chance that irrelevant instructions override a better-matched skill or influence handling of general web/search tasks unexpectedly.
The trigger keywords include generic terms such as 'web', 'search', 'api', 'user', and 'alternative', which are common across many benign requests. In a skill-routing system, such unspecific triggers can over-match and invoke this skill outside its intended domain, creating prompt-scope confusion and raising the risk of inappropriate or misleading outputs.
The trigger sentence is so broad that normal user requests containing common phrases like 'help me' could activate this skill outside its intended scope. Over-broad invocation rules can cause unintended routing, context confusion, and accidental execution of workflows that access web/search-related capabilities when the user did not explicitly request this skill.
The trigger sentences are broad, repetitive, and loosely scoped to common productivity/search terms, which increases the chance of accidental or inappropriate activation. In an agent ecosystem, overbroad activation can route unrelated user requests into this skill, causing unintended web-search behavior, confused delegation, or unsafe downstream actions based on mismatched context.
The trigger phrases include very broad everyday language such as 'Help me' and 'I need a practical workflow', which can cause the skill to be invoked unintentionally in unrelated conversations. In an agent ecosystem, overbroad activation can route user data or tasks into this skill unexpectedly, reducing user control and potentially causing unintended web/API actions.
The example trigger sentences rely on ordinary phrases like 'help me' and 'I need a practical workflow' while repeating broad requirement text rather than clearly delimiting when the skill should activate. This trains users or routing systems toward ambiguous activation boundaries, which can lead to accidental invocation and reduced trust in skill selection behavior.
The skill metadata description and overall trigger framing use very broad, common terms such as work-productivity, web, search, and via, without meaningful contextual constraints. This can cause accidental invocation in unrelated conversations, leading the agent to apply the wrong workflow, produce irrelevant outputs, or override a more appropriate skill path.
The explicit trigger keyword list contains highly generic tokens like web, search, api, user, and bug fix, which are common across many unrelated requests. In a skill-routing system, such broad matching increases the chance of false activation, confusing task selection and potentially exposing users to incorrect guidance or unintended workflow execution.
The default prompt uses a highly generic invocation phrase ('Use $work-productivity-tavily-web-workflow-helper to help me...') tied to broad terms like work-productivity, web, and workflow, which can overlap with ordinary user requests. In combination with agent routing, this raises the chance of unintended activation or over-selection of the skill, causing prompt-scope confusion and potentially exposing users to behavior they did not explicitly request.
Enabling implicit invocation without strong activation constraints allows the skill to be triggered based on vague contextual similarity rather than explicit user consent. Because this skill is framed around broad productivity and web-search workflows, accidental invocation is more likely and can lead to misrouting, unexpected tool use, or the application of this skill in contexts where it is not appropriate.
This trigger remains ambiguous because it describes a broad class of practical workflow requests without clearly limiting when the skill should be invoked. In an agent environment, ambiguous routing can misapply the skill to unrelated tasks, causing incorrect assistance, privilege overreach through unnecessary tool use, or user confusion about why web-search behavior was selected.
The README states that there are English and Chinese instruction files, indicating language-specific behavior or presentation, but it does not say how the user's preferred language is selected. Under the language policy rule, locale handling should be opt-in or clearly justified rather than implicitly fixed.
The file is labeled as a Chinese README, but substantial user-facing content such as the requirement description, workflow description, and trigger phrases remains in English. This can amount to an implicit language policy mismatch because the skill does not clearly offer language selection or explain the mixed-language behavior.
The description embeds the Chinese term "搜索" inside otherwise English guidance, which introduces an implicit language/locale assumption without offering user opt-in or explaining a region-specific need. This can conflict with language policy expectations for consistent user-facing behavior.
The natural-language description includes both English and Chinese text ("Tavily 搜索-style") without indicating whether the skill expects or enforces a particular language preference. Because no user opt-in or documented locale rationale is provided, this may conflict with organizational language-choice expectations.
The requirement description embeds the phrase "Tavily 搜索-style workflows," which introduces a specific language choice in the skill text without indicating whether users can choose their preferred language or locale. This can create an implicit language preference rather than a neutral or user-selected one.
