The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill fetches documentation and configuration files from ably.com/llms.txt and ably.com/docs. These are official vendor resources intended for LLM consumption and do not present a security risk.
[COMMAND_EXECUTION]: The documentation mentions the use of @ably/cli via npm install -g @ably/cli for testing and verification. This is a standard development tool provided by the vendor.
[CREDENTIALS_UNSAFE]: The skill proactively addresses credential safety by explicitly instructing against hardcoding API keys in client-side code and detailing how to use secure JWT or token authentication from a backend server.
[PROMPT_INJECTION]: No malicious instructions, safety bypasses, or jailbreak patterns were detected. The skill uses clear, direct instructions to ensure the agent follows official documentation rather than generating hallucinated code.

using-ably