elastic-beanstalk

Pass

Audited by Gen Agent Trust Hub on Jun 21, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [SAFE]: No malicious patterns or security risks were identified. The skill follows established cloud deployment patterns and security best practices.
  • [COMMAND_EXECUTION]: The skill instructs the agent to use standard AWS CLI commands (e.g., aws elasticbeanstalk create-application) and build tools (e.g., dotnet publish, mvn package) for its primary deployment function. All significant operations are designed to be performed with user confirmation.
  • [PROMPT_INJECTION]: Evaluated for indirect prompt injection surface: (1) Ingestion points: Repository configuration files such as package.json, Procfile, and Dockerfile are read to determine platform settings. (2) Boundary markers: Not specified. (3) Capability inventory: AWS CLI execution, archive creation, and build tool invocation. (4) Sanitization: Not specified. (5) Analysis: The processing of configuration files is a legitimate requirement for deployment automation and shows no evidence of exploitation or malicious instruction injection.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 21, 2026, 09:50 AM
Security Audit — agent-trust-hub — elastic-beanstalk