Security audit
Security checks for vulnerabilities and agentic risk
This documentation-only skill appears safe to install, though its broad auto-invocation wording may make it activate more often than intended.
Install only if you want a general helper for ontology-style productivity and typed workflow design. Be aware that its broad keywords and implicit invocation setting may cause it to appear in loosely related workflow, knowledge-graph, memory, or bug-fix requests.
The skill description uses very broad activation criteria such as generic domains and common task types, which can cause the skill to trigger for many unrelated requests. Over-broad routing can override more appropriate skills, leading to incorrect guidance, unintended authority, and expanded exposure of any downstream unsafe behavior.
The trigger keyword list includes highly generic terms like 'knowledge', 'graph', 'structured', 'creating', and 'bug fix', which overlap with a large fraction of normal user requests. In an agent routing system, such ambiguous keywords can cause persistent misselection of this skill, increasing the chance of task hijacking, degraded reliability, and inappropriate handling of sensitive workflows.
The trigger sentence is broad and generic enough that ordinary user requests about help, workflows, or hardening could unintentionally activate this skill. In an agent routing system, this can cause mis-selection, unexpected behavior, and privilege expansion if the wrong skill is invoked for unrelated tasks.
The trigger phrases are broad, generic, and closely resemble ordinary user requests for help with workflows or productivity tasks. This can cause the skill to activate unintentionally in unrelated conversations, leading to prompt-scope hijacking or inappropriate insertion of the skill's instructions into contexts where the user did not intend to invoke it.
The trigger phrases are generic and include common words like work-productivity, ontology, typed, knowledge, graph, and broad request templates, which can cause the skill to activate in contexts the user did not intend. In an agent ecosystem, unintended invocation can route tasks into the wrong workflow, causing incorrect actions, confusing outputs, or unsafe automation chaining.
The example trigger sentences use commonplace phrasing like 'help me' and 'I need a practical workflow', which provides little boundary between valid and invalid activations. This makes the skill easier to invoke accidentally or preferentially during prompt matching, weakening routing precision and potentially suppressing safer or more relevant specialized skills.
The skill description defines invocation conditions in vague, expansive language such as 'or needs practical workflow, artifact, checklist, analysis, or implementation support for this requirement,' without a clear boundary for when the requirement applies. In a multi-skill environment, this ambiguity increases the chance of overbroad activation and misclassification of user intent, which can degrade reliability and potentially expose users to unsuitable automated actions or advice.
The trigger keywords are overly broad and include generic terms such as 'knowledge', 'graph', 'structured', 'memory', 'creating', and 'bug fix', which can match many unrelated user requests. This creates skill-selection ambiguity and can cause the agent to invoke this skill outside its intended scope, leading to incorrect task routing, irrelevant guidance, or interference with more appropriate skills.
The default prompt uses very broad trigger language like work-productivity, ontology, typed, knowledge, graph, workflow, artifact, checklist, analysis, and implementation support, which can match many unrelated user requests. This increases the chance of unintended activation, causing the skill to inject guidance or workflow behavior into contexts where it was not explicitly requested.
Enabling implicit invocation without meaningful constraints allows the platform to auto-select this skill based on vague semantic overlap rather than deliberate user choice. In combination with the broad skill description, this can cause over-triggering and unintended prompt injection into benign conversations.
The trigger description uses a broad request pattern that overlaps with many normal productivity queries, making accidental invocation likely. Because this skill advertises generalized workflow, checklist, analysis, and implementation support, an overly permissive trigger increases the chance that the agent routes unrelated requests here and applies inappropriate guidance.
This is a Chinese README, but key user-facing content such as the requirement and target-user description is presented in English. That creates an implicit language behavior that is not explained or offered as an opt-in choice, which may conflict with language/locale policy expectations.
No suspicious patterns detected.
