stacks

Pass

Audited by Gen Agent Trust Hub on May 5, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill installs the @membranehq/cli package from the NPM registry. This package is the official CLI for the Membrane platform.
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute various membrane CLI commands for logging in, connecting to services, and running actions. This is the intended operation of the skill and uses the vendor's provided tools.
  • [INDIRECT_PROMPT_INJECTION]: The skill includes surfaces where untrusted or natural language data is processed.
  • Ingestion points: User-provided strings in the --intent parameter for membrane action list and the description for membrane action create (SKILL.md).
  • Boundary markers: None identified in the provided command examples.
  • Capability inventory: The skill can list, create, and execute actions on the Stacks platform using the CLI (SKILL.md).
  • Sanitization: Relies on the Membrane platform's server-side processing and validation of natural language intents.
Audit Metadata
Risk Level
SAFE
Analyzed
May 5, 2026, 10:36 PM
Security Audit — agent-trust-hub — stacks