ab-tasty

Pass

Audited by Gen Agent Trust Hub on Oct 24, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill requires the installation of the @membranehq/cli package from the public npm registry to interact with the service. This is a standard dependency for the platform provided by the vendor.
  • [COMMAND_EXECUTION]: The instructions involve executing shell commands via the Membrane CLI to manage authentication (membrane login), established connections (membrane connection ensure), and invoke API actions (membrane action run). These are controlled operations used for the skill's primary purpose.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external data from A/B Tasty campaigns and variations. This constitutes an attack surface for indirect prompt injection if the external account contains malicious instructions; however, this is a common characteristic of integration skills and is mitigated by standard LLM guardrails.
  • Ingestion points: API responses from A/B Tasty retrieved via membrane action run and membrane request (documented in SKILL.md).
  • Boundary markers: Not explicitly defined in the provided markdown instructions.
  • Capability inventory: Includes network requests and command execution via the Membrane CLI (documented in SKILL.md).
  • Sanitization: The skill relies on the Membrane platform's internal handling of API requests and schema validation.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 24, 2026, 07:49 PM