The Agent Skills Directory

[SAFE]: Detailed analysis of the provided scripts and instructions reveals no malicious patterns. The skill focuses on implementing secure webhook endpoints with proper authentication checks using the dg-token header.
[CREDENTIALS_UNSAFE]: The skill follows secure practices for credential management. It uses environment variable placeholders (DEEPGRAM_API_KEY, DEEPGRAM_API_KEY_ID) and correctly instructs users to use .env files rather than hardcoding secrets.
[EXTERNAL_DOWNLOADS]: The documentation references the hookdeck-cli, which is a legitimate tool provided by the vendor for testing webhooks locally. References to external documentation point to official Deepgram and Hookdeck resources.
[COMMAND_EXECUTION]: The README files contain curl command examples for testing purposes. These are standard instructional snippets for API interaction and do not involve unsafe execution of untrusted input.

deepgram-webhooks