The Agent Skills Directory

[SAFE]: The skill contains legitimate implementation patterns for Retell AI voice agents and telephony, focusing on platform reliability and resilience.
[INDIRECT_PROMPT_INJECTION]: The skill demonstrates processing external webhook data (transcripts) in references/resilience-code.md, which is an ingestion point for external content. The provided examples focus on latency management and caching rather than processing untrusted instructions.
Ingestion points: references/resilience-code.md (via req.body in the webhook handler)
Boundary markers: None present in the simplified architectural examples
Capability inventory: Network response generation and Redis cache interaction
Sanitization: None present in the simplified code snippets
[EXTERNAL_DOWNLOADS]: References the official retell-sdk package from the public npm registry.

retellai-reliability-patterns