Security audit
Security checks for vulnerabilities and agentic risk
This skill is a documentation helper for OpenAPI/Swagger work, with a minor risk that its broad trigger wording may invoke it too often.
Safe to install for OpenAPI/Swagger documentation assistance. Be aware that the skill may activate on broad API or developer-experience wording; invoke it explicitly when you want API documentation help and ignore it for unrelated backend engineering tasks.
The trigger sentence is so generic that it can match ordinary user phrasing rather than a clear, deliberate invocation of this specific skill. That creates unintended activation risk, where the agent may route unrelated requests into this skill and produce documentation-oriented behavior in contexts the user did not explicitly request.
The trigger phrases are broad and partially templated, making accidental or overly eager invocation more likely when a user mentions common terms like OpenAPI, Swagger, or practical workflow help. In an agent ecosystem, unintended activation can route user requests into the wrong skill, causing context confusion, unsafe tool use, or bypass of more appropriate safeguards.
The trigger phrases are broad and can activate the skill from generic requests containing common API-documentation terms, increasing the chance of unintended invocation. In an agentic system, overbroad activation can route user input into a workflow the user did not explicitly choose, causing prompt/skill confusion and potentially exposing internal context or producing actions outside user intent.
The skill description and activation guidance are broad enough to match many general software or API-related requests, which can cause the agent to invoke this skill when the user did not specifically want OpenAPI documentation help. Over-broad routing is dangerous because it can override more appropriate skills, produce irrelevant outputs, or steer handling of requests into the wrong workflow, reducing reliability and potentially affecting security-sensitive requests that mention APIs only incidentally.
The trigger keywords include ambiguous terms such as 'software-and-data', 'openapi', 'swagger', 'api documentation', 'rest api', and 'developer experience' without enough qualifying context. This increases the chance of accidental activation on unrelated prompts, which can misroute the conversation and cause the system to apply this skill's instructions in contexts where they do not fit.
The skill trigger description includes broad terms such as software-and-data, openapi, swagger, api documentation, and rest api without clear scoping logic or exclusion criteria. This can cause the skill to activate during general technical conversations, leading to unnecessary context injection, user confusion, or unintended steering of the agent's behavior.
The keyword "developer experience" is overly broad and applies to many unrelated engineering discussions. In a skill-routing system, this can cause misfires into this skill for requests that have nothing to do with REST API documentation, which degrades routing integrity and may expose irrelevant instructions or outputs.
The example trigger phrases use generic openings like "Help me" and "I need a practical workflow" without any exclusion rules or precise task boundaries. Such vague exemplars encourage broad pattern matching and increase the chance that unrelated requests will activate the skill, weakening separation between skills and creating avoidable prompt-scope confusion.
The default_prompt uses a broad, vague invocation phrase that can cause the skill to be triggered in situations beyond its intended scope. Because implicit invocation is enabled, this ambiguity increases the chance of the agent activating on loosely related requests and influencing outputs or workflows without clear user intent.
The trigger examples are ambiguous and do not clearly constrain when the skill should activate, which can cause overbroad routing and misfires. In an agentic system, ambiguous triggers weaken user-intent boundaries and can lead to the wrong workflow being applied, reducing reliability and potentially exposing downstream tools or prompts to irrelevant inputs.
The README advertises separate English and Chinese documentation files, but it does not state how language is selected or whether users can choose their preferred locale. This can create ambiguity around locale behavior and may conflict with organizational expectations that language choice be explicit rather than assumed.
The file is named as a Chinese README, but key user-facing instructional content such as the demand description, workflow description, keywords, and trigger phrases remains in English. This creates an implicit language constraint/mismatch for users expecting a Chinese-localized skill description, without any opt-in or language-selection guidance.
No suspicious patterns detected.
