Security audit
Security checks for vulnerabilities and agentic risk
This is a documentation-only workflow helper with no executable code, credentials, persistence, or hidden data access, though its activation wording is overly broad.
This skill appears safe to install from a security perspective. Users should be aware that its broad trigger terms and implicit invocation setting may make it appear in unrelated workflow or knowledge-graph conversations, so maintainers should narrow the activation criteria for better routing.
The skill description and use criteria are extremely broad, covering generic requests like workflow, checklist, analysis, and implementation support across multiple adjacent concepts. This can cause the skill to activate in situations outside its intended domain, leading to unintended routing, prompt-scope capture, and interference with more appropriate or safer skills.
The trigger keywords include highly generic terms such as 'typed', 'knowledge', 'graph', 'structured', 'memory', 'creating', and 'bug fix' without contextual constraints. These words appear in many unrelated requests, so the skill may be invoked accidentally and overshadow more relevant skills, increasing the risk of incorrect behavior or prompt-routing abuse.
The trigger sentence is broad enough to match ordinary user language, which can cause the skill to activate outside its intended context. In an agent ecosystem, overbroad activation can route unrelated requests through this skill, causing inappropriate behavior, confusing outputs, or interference with safer or more relevant skills.
The invocation guidance is ambiguous because the examples are long, awkward requirement text instead of a precise trigger contract. This increases the chance that downstream systems or users misunderstand when the skill should run, which can create misrouting, prompt collisions with other skills, and unreliable behavior in multi-skill agent environments.
The trigger sentence is overly broad and reads like ordinary user language rather than a narrowly scoped invocation rule. In an agent-routing context, this can cause accidental activation on unrelated requests, leading the wrong skill to take over handling, produce irrelevant workflow artifacts, or interfere with safer/more appropriate skills.
The trigger phrases are broad, generic, and partially templated around common requests like asking for practical help or workflows. This can cause the skill to activate for unrelated user intents, increasing the chance of unintended routing, confused execution, or inappropriate application of the skill in contexts where it was not meant to be used.
The example trigger sentences are themselves vague and repetitive, and they normalize activation on broad phrases rather than clarifying safe boundaries. This reinforces over-triggering behavior and makes it easier for unrelated user input to match the skill, reducing routing precision and potentially exposing users to the wrong workflow guidance.
The invocation description says to use the skill whenever a user asks for broad concepts like work-productivity, ontology, typed, knowledge, or graph, or for generic artifacts such as workflows, checklists, analysis, or implementation support. Because these categories are expansive and lack exclusion criteria, the skill may activate in many contexts outside its intended scope, leading to misrouting and over-application of the skill.
The trigger list includes very generic terms such as "typed", "knowledge", "graph", "structured", and "creating", which can match many unrelated user requests. This can cause unintended skill invocation, routing conversations into an ontology/workflow helper when the user did not ask for it, increasing the chance of irrelevant guidance, prompt-surface expansion, or interference with more appropriate skills.
The default prompt is overly broad and vague, using generic productivity and workflow language that can match many ordinary user requests unrelated to this specific skill. In combination with a named skill reference and open-ended wording, this increases the chance of unintended routing or invocation, causing the agent to apply this skill outside its intended scope and potentially produce misleading or unsafe assistance.
Enabling implicit invocation without meaningful activation constraints allows the platform to trigger this skill automatically based on weak or ambiguous similarity signals. Because the skill description is broad and covers common workplace tasks, this can lead to over-invocation, incorrect tool selection, and accidental exposure of user context to a skill that was not explicitly requested.
The activation guidance is ambiguous and lacks clear boundaries, so the routing system may select this skill for loosely related requests involving work, knowledge, graphs, or practical help. That increases the chance of prompt/skill misselection, which can degrade reliability and potentially bypass more appropriate controls or specialized handling in other skills.
The file is labeled as a Chinese README, but substantial instructional content and trigger text remain in English, with no note about language preference or user opt-in. This can amount to an implicit language policy choice rather than explicitly allowing the user to select their preferred language.
No suspicious patterns detected.
