Security audit
Security checks for vulnerabilities and agentic risk
This is a documentation-only workflow helper with no executable or privileged behavior, though its broad triggers could make it activate unexpectedly.
Install only if you want a general PollyReach-style workflow helper. Prefer explicit invocation or tighten the trigger keywords before enabling implicit invocation, because the current generic terms may route unrelated requests into this skill.
The trigger sentences are broad, templated, and only loosely tied to a narrowly scoped task, which increases the chance the skill activates for unrelated user requests. In an agent ecosystem, unintended activation can route conversations into the wrong workflow, causing mis-execution, confusion, or unsafe automation in contexts the skill was not designed to handle.
The trigger keywords and example phrases are extremely broad, including common terms such as 'every', 'phone', 'number', 'ability', 'things', and 'done'. This can cause the skill to activate on unrelated user requests, leading to unintended routing, prompt hijacking of normal workflows, or accidental exposure of users to behavior they did not request.
The skill description is scoped to very common terms and broad task shapes such as 'practical workflow, artifact, checklist, analysis, or implementation support,' which can match many unrelated user requests. This creates a real risk of unintended activation, causing the wrong skill to intercept conversations and potentially steer outputs inappropriately or expose users to irrelevant or unsafe workflow guidance.
The trigger keywords include generic everyday terms like 'every,' 'phone,' 'number,' 'ability,' 'things,' and 'done,' which are likely to appear in a wide range of normal conversations. Such unconstrained keywords greatly increase accidental invocation and skill-routing errors, making the skill dangerous in context because it is positioned as a broad workflow helper rather than a tightly bounded domain tool.
The trigger list includes very broad everyday terms such as "every", "phone", "number", "ability", "things", and "done", which are likely to match many unrelated user requests. This can cause the skill to be invoked outside its intended scope, leading to incorrect routing, confusing outputs, and increased attack surface if sensitive or privileged workflows are attached to the skill.
The trigger examples are broad, fragmentary, and include common terms like "help me" and generic workflow language that could cause the skill to activate for unrelated user requests. Over-broad activation increases the chance of unintended routing, context capture, or execution of this skill when a more appropriate skill should handle the task.
The keyword list includes highly generic terms such as 'every', 'phone', 'number', 'ability', 'things', and 'done', which can match a wide range of unrelated prompts. This creates a prompt-routing vulnerability where the skill may be invoked outside its intended work-productivity context, increasing the risk of irrelevant or unsafe behavior due to scope confusion.
This file is explicitly a Chinese-language README variant, but the content does not state that language selection is optional or user-chosen, nor does it justify a locale restriction. Under the policy, language or locale constraints should be opt-in or clearly documented as region-specific.
The example trigger sentences begin with highly generic phrases like 'Help me' and 'I need a practical workflow,' which can train or bias invocation systems toward activating on ordinary assistance requests. While less severe than the keyword list itself, this still broadens the effective trigger surface and increases the chance that unrelated tasks are misrouted to this skill.
The description says to use the skill when users mention broad concepts or need "practical workflow, artifact, checklist, analysis, or implementation support," but it does not clearly define exclusions or precise activation criteria. This ambiguity increases the chance of accidental invocation and misapplication, especially when combined with the overly broad trigger keywords later in the file.
The default prompt contains highly generic trigger terms and a broad invocation pattern, which can cause the skill to activate in ordinary user conversations unrelated to the intended workflow. Because implicit invocation is enabled, this increases the chance of unintended routing, prompt confusion, or accidental execution of skill-specific behavior in contexts where the user did not mean to call this skill.
The invocation section does not define strong boundaries for when the skill should or should not run, and the listed keywords are generic enough to overlap with ordinary conversation. In an agent ecosystem, this can lead to accidental invocation, misrouting, and expanded exposure of user context to a skill that was not intentionally selected.
No suspicious patterns detected.
