Security audit
Security checks for vulnerabilities and agentic risk
This is a documentation-only workflow helper with overly broad activation wording but no evidence of malicious behavior, persistence, credential access, or hidden execution.
Install only if you want a broad workflow-planning helper for ontology or structured productivity tasks. Be aware that its implicit invocation and generic triggers may cause it to appear on loosely related requests; explicit invocation or narrower triggers would make selection more predictable.
The skill description and use conditions are broad enough to activate on generic requests for workflows, analysis, implementation support, or common terms like knowledge and graph. This can cause the wrong skill to be selected for unrelated tasks, leading to confused delegation, unsafe context mixing, or bypass of more appropriate specialized skills.
The trigger keywords include highly generic terms such as knowledge, graph, structured, memory, creating, and bug fix without any gating conditions. In an agent ecosystem, this increases accidental invocation and prompt-routing collisions, which can misapply the skill to unrelated user intents and reduce safety and reliability of task handling.
The trigger examples include broad natural-language phrases like 'Help me' and 'I need a practical workflow', which are common in ordinary user requests and can cause unintended skill invocation. In an agent ecosystem, ambiguous triggers can route unrelated prompts into this skill, creating prompt-scope confusion and increasing the chance that the skill acts on requests it was not specifically intended to handle.
The trigger phrases are broad, generic, and partially mirror ordinary user requests for workflows, bug fixing, and hardening. This increases the chance that the skill is invoked unintentionally in unrelated contexts, which can cause inappropriate prompt routing, confusion, or application of the wrong workflow to sensitive tasks.
The trigger phrases are generic enough to match ordinary requests for help, workflows, or bug fixing, which can cause the skill to activate in contexts the user did not intend. In an agent ecosystem, over-broad invocation increases the chance of prompt-routing mistakes, inappropriate tool use, or unintended disclosure of context to a mismatched skill.
The top-level description says to use the skill whenever a user asks for broad categories like work-productivity, ontology, typed, knowledge, graph, or practical workflow/checklist/analysis support. This scope is so broad that it can match a large portion of normal agent requests, leading to accidental invocation and reduced reliability of skill selection.
The trigger list includes very common and generic terms such as "typed", "knowledge", "graph", "structured", "creating", and "bug fix", which can cause the skill to activate for many unrelated requests. Over-broad activation increases the chance of unintended routing, causing users to receive irrelevant workflow guidance or have this skill override more appropriate, narrower skills.
The manifest enables implicit invocation while pairing it with a vague, marketing-style default prompt that does not narrowly constrain what the skill should do. This increases the chance the platform auto-selects the skill for loosely related requests, causing overbroad activation, unintended data exposure to the skill context, or unsafe workflow execution without clear user intent.
The invocation guidance is malformed and ambiguous: it mixes long requirement text into trigger examples and does not clearly define when the skill should or should not activate. This can lead to overbroad matching, accidental invocation, or inconsistent agent behavior, especially because the skill is framed for general workflow assistance and could be selected for loosely related tasks.
This file presents the skill documentation entirely in Simplified Chinese, but does not state that language selection is optional or explain why the locale is required. For a general-purpose skill aimed at broad AI-agent users, this can be a natural-language policy issue unless the locale restriction is justified or paired with an explicit language choice.
