The Agent Skills Directory

[CREDENTIALS_UNSAFE]: The skill demonstrates secure credential management by using environment variables via process.env.DEEPGRAM_API_KEY and the dotenv library. No hardcoded secrets were found.
[EXTERNAL_DOWNLOADS]: The skill references official Deepgram documentation, OpenAPI/AsyncAPI specifications, and an example audio file hosted on a vendor-controlled domain (dpgr.am). It also mentions adding central product skills using npx skills add deepgram/skills, which is the standard distribution mechanism for this vendor.
[SAFE]: No prompt injection, data exfiltration, obfuscation, or malicious persistence mechanisms were detected. The technical instructions are consistent with the stated purpose of assisting with audio intelligence analytics using the Deepgram SDK.

deepgram-js-audio-intelligence