retellai-cost-tuning

Pass

Audited by Gen Agent Trust Hub on Sep 9, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill uses the official Retell AI SDK and API endpoints (api.retellai.com) for telephony cost management and agent configuration.
  • [DATA_EXPOSURE]: Authentication is handled via the RETELL_API_KEY environment variable, which is a recommended practice for secure credential management.
  • [COMMAND_EXECUTION]: Provides bash scripts using curl and jq to monitor call durations and costs, and to update agent settings via the official API.
  • [INDIRECT_PROMPT_INJECTION]:
  • Ingestion points: Fetches call history data from the Retell AI API in references/cost-optimization.md.
  • Boundary markers: None; data is processed using jq for statistical analysis.
  • Capability inventory: Includes capabilities to modify agent configuration (PATCH requests) via curl in references/cost-optimization.md.
  • Sanitization: None; however, the skill specifically targets numeric fields (duration, cost) for analysis, minimizing exposure to unstructured data that could contain injection patterns.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 9, 2026, 03:46 AM
Security Audit — agent-trust-hub — retellai-cost-tuning