amazon-sagemaker
Pass
Audited by Gen Agent Trust Hub on Apr 30, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: Installs the
@membranehq/clipackage from the public npm registry. This is a vendor-owned CLI tool required for interaction with the Membrane platform. - [COMMAND_EXECUTION]: Uses
membranecommands to authenticate users, search for available Sagemaker actions, and execute machine learning tasks. - [PROMPT_INJECTION]: The skill contains an indirect prompt injection surface by processing data retrieved from Sagemaker actions.
- Ingestion points: Output from
membrane action runin SKILL.md. - Boundary markers: None present in the skill instructions.
- Capability inventory: Ability to create and run Sagemaker actions, models, and jobs in SKILL.md.
- Sanitization: No specific sanitization or validation of external data is specified.
- [SAFE]: Implements secure credential handling by using a centralized connection manager, preventing the need for local API keys or tokens.
Audit Metadata