amazon-alexa

Pass

Audited by Gen Agent Trust Hub on Aug 6, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill recommends installing official command-line utilities including the Amazon Alexa Skills Kit CLI (ask-cli) and the Amazon Web Services CLI (awscli) for project management and deployment.
  • [EXTERNAL_DOWNLOADS]: Uses the official ask new command to download project templates from Amazon's repositories.
  • [DATA_EXFILTRATION]: Contains instructions for standard developer credential setup via aws configure and ask configure to manage local access to AWS and Alexa services.
  • [PROMPT_INJECTION]: The skill architecture facilitates the processing of external user voice input through a chat intent that interacts with an LLM backend.
  • Ingestion points: User input is captured via the query slot in the ChatIntent handler within SKILL.md.
  • Boundary markers: None; the skill appends user input directly to the message history for the Claude API.
  • Capability inventory: The backend handler (chat_handler) calls the Claude API, manages persistent user history in Amazon DynamoDB, and provides potential control over IoT devices through the Smart Home API.
  • Sanitization: No explicit input sanitization or escaping is implemented before interpolating voice text into the LLM prompt.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 6, 2026, 01:39 AM
Security Audit — agent-trust-hub — amazon-alexa