Security audit
Security checks for vulnerabilities and agentic risk
This is a documentation-only workflow helper with no hidden code or privileged behavior, though its activation terms are too broad.
Installers should be aware that this skill may trigger on broad security or workflow language. It appears safe as a documentation-only helper, but its activation terms should ideally be narrowed to explicit SkillScan-style workflow requests.
The trigger examples are phrased as ordinary help requests and broad workflow language, which raises the chance that the skill activates when a user did not explicitly intend to invoke it. In an agent ecosystem, unintended activation can redirect task flow, cause incorrect tool selection, or introduce unsafe behavior inherited from the skill into unrelated contexts.
The trigger phrases are broad, natural-language prompts that can easily match ordinary user requests unrelated to this skill. That creates an over-activation risk where the skill may inject its workflow or instructions into unintended contexts, potentially bypassing user intent and interfering with other skills or safety boundaries.
The skill description says it should be used when a user asks for terms like 'security', 'gate', or 'every', which are extremely broad and likely to appear in many unrelated conversations. This can cause unintended activation or routing, increasing the chance that the skill injects irrelevant workflow instructions into unrelated tasks and expands its authority beyond the intended scope.
The keyword list includes vague terms like 'security', 'gate', 'every', 'must', 'pass', 'before', 'activate', and 'bug fix', which overlap heavily with normal user language. In an agent ecosystem, such generic triggers can make the skill activate promiscuously, creating prompt-scope confusion, misrouting, or accidental execution of this skill in conversations that never intended to invoke it.
The trigger keyword list includes very broad everyday terms such as "security", "every", "must", "pass", and "before", which can cause the skill to activate in many unrelated conversations. Overbroad activation is dangerous because it can unexpectedly inject this skill’s workflow into other tasks, confusing system behavior, bypassing intended routing boundaries, and increasing the chance that users receive inappropriate or misleading guidance.
The keyword list includes generic terms such as 'security', 'gate', 'every', 'must', 'pass', 'before', and 'activate', which are common in normal conversation and likely to collide with unrelated requests. This can cause accidental routing to the skill, expanding its influence beyond intended use and potentially interfering with security-sensitive workflows.
The keyword list includes highly generic terms like 'security', 'must', 'pass', 'before', and 'bug fix', which are common across many benign conversations. This ambiguous activation scope increases the chance of accidental invocation and prompt-scope collisions, making agent behavior less predictable and easier to manipulate through incidental wording.
The file specifies separate English and Chinese documentation variants but does not tell the user they can choose their preferred language or locale. In a locale-specific README, omitting an explicit language choice can amount to forcing a language context without opt-in.
The example trigger sentences are malformed, incomplete, and not framed as precise activation patterns, so they do not define a safe boundary for when the skill should run. Poor examples reinforce overbroad matching and make it more likely that downstream systems or authors copy unsafe trigger conventions that activate on incidental text.
The description says the skill should be used when a user mentions broad categories like work-productivity, skillscan, security, gate, or every, or whenever they need practical workflow or analysis support for the requirement. This is dangerously underspecified because it greatly expands the activation surface, making accidental invocation likely and weakening separation between this skill and unrelated skills or default assistant behavior.
The example trigger phrases are affirmative examples only and do not establish clear limits or counterexamples, so they fail to constrain the activation scope. In combination with the broad keywords, this can train or encourage overly permissive matching logic, increasing accidental routing rather than reflecting deliberate user intent.
The default prompt is phrased as a broad natural-language trigger tied to common terms like work-productivity and security-related workflow help, while implicit invocation is enabled. This increases the chance the skill is auto-invoked in unintended contexts, causing prompt-scope expansion, accidental routing of sensitive user requests into this skill, or abuse through trigger collisions with unrelated workflows.
The trigger sentences and keywords are broad enough to match common phrases such as 'security', 'workflow', 'help me', or 'I need a practical workflow', which can cause the skill to activate in contexts far beyond its intended scope. Over-broad activation increases the chance of prompt/skill routing hijack, unintended invocation, and accidental exposure of this skill's behavior in unrelated conversations.
