Security audit
Security checks for vulnerabilities and agentic risk
This is a documentation-only workflow helper with overly broad auto-invocation wording, but no executable code, persistence, credential handling, or hidden data flow was found.
Installers should be aware that this skill may auto-activate too broadly for ordinary web, search, API, or workflow requests. It appears safe as a guidance-only helper, but narrowing invocation terms would improve reliability.
The trigger sentences are written as broad natural-language requests that could match many ordinary user prompts unrelated to this exact skill. In an agent routing system, that can cause unintended skill activation, leading to incorrect tool selection, irrelevant workflow execution, and increased exposure to downstream web/search capabilities when they were not actually needed.
The trigger phrases are broad, natural-language prompts that could match ordinary user requests and cause the skill to activate unintentionally. In an agent ecosystem, accidental invocation can route user tasks into an unintended workflow, leading to incorrect tool usage, confusing behavior, or unreviewed web-search actions that the user did not explicitly request.
The documented trigger phrases are broad, generic, and closely resemble normal user requests for help or workflows. This can cause the skill to activate unintentionally in unrelated contexts, leading to prompt hijacking of general conversations, inappropriate tool/workflow invocation, or unexpected processing of user data under the skill's behavior.
The skill description is extremely broad and includes common terms like 'web', 'search', and 'via', which can cause the skill to activate for many unrelated requests. This increases the chance of unintended routing, where users receive irrelevant or misleading workflow guidance and other more appropriate skills are bypassed.
The trigger list contains highly generic keywords such as 'web', 'via', 'api', and 'user', which are likely to appear in ordinary conversation. An attacker or even normal user traffic could unintentionally trigger this skill, causing misrouting, prompt-surface expansion, and reduced reliability of agent behavior.
The skill description defines applicability in very broad terms and does not clearly bound when the skill should or should not be used. Ambiguous routing criteria can cause the orchestrator to select this skill for loosely related requests, creating misexecution risk and making it easier for adversarial or irrelevant prompts to steer the agent into the wrong workflow.
The trigger keyword list includes very broad, common terms such as 'web', 'search', 'api', and 'user', which can cause the skill to activate in many unrelated conversations. Over-broad activation increases the chance of prompt/context injection into sessions where the skill is not appropriate, leading to unintended guidance, workflow interference, or accidental exposure of unrelated user context to the skill logic.
The default prompt is phrased broadly enough that many normal productivity or troubleshooting requests could match it, making unintended routing more likely. When combined with implicit invocation, this broad wording can cause overreach, where the skill is injected into conversations beyond its intended domain.
The skill enables implicit invocation without any narrowly defined trigger constraints, which means it can be auto-selected for broad classes of ordinary user requests. In a workflow/search helper, this increases the chance of unintended activation, prompt-surface expansion, and misuse in contexts the user did not explicitly request.
The keyword list contains very generic terms such as 'web', 'via', 'user', and 'bug fix', which are likely to appear in many unrelated requests. Overbroad invocation keywords can make the skill activate outside its intended scope, reducing routing accuracy and potentially invoking search-oriented behavior in contexts where another safer or more appropriate skill should handle the request.
The document is presented as an English README but includes Chinese text such as "搜索-style" without explaining language selection or locale behavior. This can violate language/locale expectations because the skill does not state that users can choose their preferred language or that the mixed-language content is intentional for a specific audience.
The file is presented as a Chinese README but contains substantial English instructional content and English trigger phrases, while not stating whether the skill expects or enforces a particular interaction language. This can create a language/locale policy issue because users are not given an explicit language choice or opt-in.
The natural-language description includes the Chinese term '搜索-style' inside otherwise English instructions, which imposes a locale/language choice without explanation or user opt-in. This can conflict with language consistency expectations for users who did not request mixed-language output.
The manifest text includes the term "搜索-style" in otherwise English-facing descriptions, which can impose a language expectation without user opt-in. There is no indication that the skill is region-specific or that users may choose their preferred language/locale.
