Security audit
Security checks for vulnerabilities and agentic risk
This is a documentation-help skill for OpenAPI/Swagger work, with no executable code or hidden data access, though its triggers are broader than ideal.
Installers should be aware that this skill may trigger on broad API or developer-experience wording. It appears safe as a local documentation assistant, but narrower trigger wording would make invocation more predictable.
The trigger sentence begins with a highly generic phrase ('Help me ...'), which can cause the skill to activate for broad, ordinary user requests that are not specifically about OpenAPI or Swagger work. Over-broad activation increases the chance of incorrect routing, prompt-context leakage into unrelated tasks, and untrusted invocation of the skill in situations the user did not intend.
The trigger sentences are highly generic and include broad phrasing such as 'Help me' and 'I need a practical workflow,' which can cause the skill to activate for loosely related requests. In an agent environment, ambiguous invocation increases the chance of misrouting user tasks, unexpected skill execution, and accidental application of domain-specific instructions outside the intended OpenAPI documentation context.
The trigger phrases are highly generic (for example, broad requests for help or practical workflows) and can cause the skill to activate in contexts the user did not explicitly intend. In an agent environment, unintended invocation can route unrelated prompts into this skill, leading to incorrect handling, prompt confusion, or disclosure of context to a skill that was not meant to run.
The skill description uses broad trigger terms like 'software-and-data', 'api documentation', and 'rest api' without meaningful narrowing conditions. This can cause the skill to activate for loosely related requests, increasing the chance of misrouting user queries, overshadowing more appropriate skills, or being invoked in contexts it was not designed to handle.
The keyword trigger list contains generic phrases that are common across many unrelated engineering tasks, but it does not specify qualifiers or exclusions. In a skill-routing system, this broad matching can lead to over-triggering and inappropriate delegation, which degrades reliability and may expose downstream workflows to unintended inputs.
The skill description and usage conditions are broad enough to match many normal software-help requests, which can cause the skill to activate outside a narrowly intended OpenAPI-documentation context. Over-broad routing increases the chance that a user is steered into an irrelevant workflow, receives misplaced guidance, or that more specialized safeguards from a better-matched skill are bypassed.
The keyword list includes generic terms such as software-and-data and developer experience that are not specific to OpenAPI documentation work. Generic triggers can cause accidental invocation on unrelated conversations, broadening the skill's effective scope and making prompt-routing behavior less predictable.
The skill enables implicit invocation (allow_implicit_invocation: true) without any visible trigger constraints, narrowing conditions, or approval gates. That can cause the agent to invoke this skill automatically on loosely related prompts, expanding the attack surface for prompt injection, unintended data flow into the skill, or surprising autonomous behavior.
The trigger sentence ('I need a practical workflow for ...') is ambiguous and incomplete, so it does not define a clear activation boundary for when the skill should run. Ambiguous triggers can cause accidental invocation on loosely related workflow requests, reducing user control and increasing the risk of the skill being applied to the wrong task context.
This is labeled as a Chinese README, but key user-facing sections such as the demand description, workflow description, keywords, and trigger phrases are written in English. That can impose an implicit language choice on users without opt-in or explanation, which may conflict with language/locale policy expectations.
The example triggers are malformed and effectively reduce to broad prompts like "Help me" and "I need a practical workflow," which overlap with common user phrasing. Because they do not show clear scope constraints, they may encourage over-triggering rather than demonstrating precise invocation conditions.
The example trigger phrases use vague, everyday invocation wording that does little to constrain when the skill should be selected. Although lower impact than the broad description and keyword list, such examples can still reinforce over-selection and increase accidental activation frequency.
