Security audit
Security checks for vulnerabilities and agentic risk
This is a documentation-only workflow helper with broad activation wording but no executable code, credential access, persistence, or data export.
Installers should be aware this skill may activate too broadly for general developer prompts because implicit invocation is enabled and common words like run, api, cli, and issue are listed as triggers. It appears safe as a guidance-only skill, but tighter activation wording would reduce confusion.
The skill description and usage guidance are broad enough to match many ordinary software-help requests, which increases the chance of unintended invocation. Over-broad routing can cause the wrong skill to activate, leading to unsafe assumptions, irrelevant actions, or bypass of more specific skills with tighter safeguards.
The trigger keywords include highly generic terms such as 'run', 'api', 'cli', and 'issue', which are common across many unrelated tasks. This makes accidental activation likely and can misroute user requests into a skill that may generate repository or workflow guidance in contexts where it is not appropriate.
The trigger sentence is so broad and generic that it can cause the skill to activate for ordinary help requests that are only loosely related to GitHub-style workflows. Over-broad activation increases the chance of unintended routing, where users receive this skill in contexts it was not designed for, potentially bypassing more appropriate or safer specialized handling.
The trigger sentences are extremely broad and partially templated, which can cause unintended activation for loosely related requests. In an agent ecosystem, overbroad routing increases the chance that the wrong skill is invoked, leading to inappropriate actions, misleading guidance, or execution in a context the user did not intend.
The trigger phrases are broad enough to match ordinary user requests about GitHub, issues, APIs, or bug fixes, which can cause the skill to activate outside its intended scope. In an agent system, overbroad routing can expose users to unintended workflows, increase prompt-surface for adversarial content, and lead to incorrect or unsafe task execution.
The example trigger phrases are ambiguous, natural-language sentences that resemble everyday requests rather than distinct invocation patterns. This trains or encourages routing on weak signals, increasing the risk that the skill activates in unrelated conversations and produces mismatched or unsafe assistance.
The trigger keywords include very broad and common terms such as run, api, issue, and github, which can cause this skill to activate in many unrelated conversations. Overbroad activation increases the chance that the wrong skill handles a request, leading to confused execution paths, incorrect guidance, or unsafe actions being suggested in contexts the skill was not meant to cover.
The example trigger sentences are generic requests like 'Help me' and 'I need a practical workflow', which do not clearly constrain the intended use case. These examples can train or bias routing toward invoking this skill for routine developer-language prompts even when the user did not ask for GitHub-style workflow assistance specifically.
The default prompt uses very broad natural-language wording such as software, data, GitHub, practical workflow, analysis, and implementation support, which overlaps heavily with normal developer conversations. In combination with implicit invocation, this creates a significant risk of accidental triggering, causing the skill to steer responses or operational behavior outside the user's explicit consent.
Implicit invocation is enabled globally without any narrowly defined trigger scope, so the skill may activate during ordinary conversations that merely resemble its broad domain. Because this skill targets common GitHub and developer-helper workflows, unintended activation could cause the agent to inject workflow behavior, prompts, or actions when the user did not explicitly request this skill.
The invocation guidance lacks specificity about when the skill should and should not be used, which makes accidental or overly aggressive invocation more likely. In an agent ecosystem, ambiguous routing criteria can lead to prompt-surface expansion, misapplication of workflows, and reduced assurance that the selected skill matches the user's actual need.
The file presents separate English and Chinese documentation variants, but the user-facing guide labels one set as English and another as Chinese without any stated user choice or default-language policy. This can be read as a locale/language handling policy that is undocumented from the user's perspective.
This file is presented as a Chinese README, but key operational text such as the workflow description and trigger phrases is written in English only. That creates an implicit language constraint for users without stating a language option or justifying the locale choice.
