The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill references documentation and API endpoints from assemblyai.com, which is a well-known service provider for speech-to-text functionality. These references are for documentation purposes and do not involve untrusted code execution.- [CREDENTIALS_UNSAFE]: The instructions explicitly mandate the use of a secret manager for ASSEMBLYAI_API_KEY and strictly forbid printing, committing, or exposing the key in URLs, adhering to security best practices.- [DATA_EXFILTRATION]: The skill defines clear boundaries for output, specifying that sensitive content such as transcript text, audio, and customer-derived data should be excluded from reporting.- [INDIRECT_PROMPT_INJECTION]: While the skill processes external fixtures, it includes specific validation and sanitization steps (e.g., content-free identifiers, deterministic validation) to mitigate risks associated with untrusted data ingestion.

assemblyai-hello-world