As a rule we assume any data coming from any user (even good users) is dirty, filthy, data. That is, nothing touches our databases without being properly cleaned and quoted. We do frequent internal security audits to ensure things like SQL injections do nothing more than be silently rejected by our system.
Now, to answer your question more directly, I'll assume you mean the "find-pkey" example in the web API. The general rules of internet security apply to that script. That is, if a hacker has access to your box (let's say through a vulnerability in Apache) then they will have access to your LimeLM API key and they can do anything they want with your account. We monitor for odd behavior like this on our end.
If your server is secure (you're using the latest patched versions of all your software) then the script will be fine. That is, there are no vulnerabilities in the script.