Security audit
Security checks for vulnerabilities and agentic risk
This is a documentation-only workflow helper with overly broad activation wording but no hidden code, persistence, credential access, or destructive behavior.
Installers should treat this as a low-risk helper, but should tighten the activation criteria before broad deployment so it only triggers for explicit AdMapix or closely related raw-data workflow requests.
The trigger phrase is so broad and unnatural that it can match loosely related user requests and cause the skill to activate outside its intended scope. In an agent ecosystem, overly broad activation increases the chance of misrouting tasks, injecting irrelevant instructions into conversations, and overriding more appropriate skills, which can degrade safety and reliability.
The skill description and activation guidance are broad enough to match many unrelated requests, which can cause accidental invocation outside the intended AdMapix-style workflow context. Over-broad routing is dangerous because it can steer users into irrelevant or lower-safety guidance, increasing the chance of incorrect actions, confusion, or misuse of agent capabilities.
The keyword list includes highly ambiguous terms such as 'raw', 'data', 'layer', 'apps', and 'bug fix', which are common across many benign and unrelated requests. In an agent system, such generic triggers can cause frequent false activations, misrouting user requests and potentially bypassing better-matched specialized skills or safety guardrails.
The skill file is entirely in Chinese and does not indicate any language negotiation or user-choice mechanism, which can force a language mismatch between the skill and the user. In practice, this can cause the agent or maintainer to misunderstand constraints, triggers, or safety steps, increasing the risk of incorrect execution, missed safeguards, and unsafe outputs.
The trigger sentence begins with a generic helper phrase ('Help me ...') that is common in normal user conversations, which can cause the skill to activate when the user did not actually intend to invoke this specific workflow. In agent routing systems, overly broad activation language can misroute tasks, override more appropriate skills, and increase the chance of unintended processing of unrelated user inputs.
The activation description does not clearly define boundaries for use, so the agent may invoke this skill for general software, data, or bug-fix requests that are unrelated to the intended AdMapix-style workflow. Ambiguous scope is dangerous because it expands the skill's authority surface and can lead to accidental takeover of user flows, context confusion, and unsafe application of instructions in unintended domains.
The trigger phrases are broad, repetitive, and partially generic, which increases the chance the skill will activate for loosely related requests rather than explicit user intent. In an agent ecosystem, unintended invocation can route user data or actions through the wrong workflow, causing mis-execution, confusion, or unsafe automation paths.
The example trigger uses a very general opener ('Help me') and then embeds a vague requirement summary rather than a constrained task pattern. Example triggers shape real routing behavior and author expectations; if they are underspecified, they reinforce over-activation and make it easier for unrelated requests to match this skill.
The description states the skill should be used for broad categories such as 'software-and-data, admapix, raw, data, layer' or whenever practical workflow help is needed, but it does not clearly define boundaries. Ambiguous activation criteria increase the chance of misrouting unrelated tasks into this skill, which can produce incorrect guidance or expose users to workflows not intended for their request.
The trigger keywords are overly broad, including generic terms like 'raw', 'data', 'layer', 'apps', and 'bug fix'. This can cause the skill to activate for unrelated requests, leading to scope confusion, unintended instruction injection opportunities, or accidental application of domain-specific workflows in the wrong context.
The skill enables implicit invocation while using a very broad description and default prompt that can match many loosely related software, data, workflow, checklist, analysis, or implementation requests. This increases the chance the agent invokes the skill unexpectedly, causing prompt-scope confusion, unintended data exposure to the skill context, or unsafe delegation to a capability the user did not explicitly request.
The trigger sentence is too vague about what exact conditions should cause activation, so the routing layer may match loosely related requests that only mention a 'practical workflow.' This ambiguity is dangerous because it reduces precision in skill selection and can lead to inconsistent or inappropriate handling of user requests, especially in environments with many overlapping software-help skills.
This README is a Chinese-language file, but substantial user-facing content including core description and trigger phrases is written in English. That can create an implicit language constraint or inconsistent locale experience without explicit user opt-in or explanation.
