flipando

Pass

Audited by Gen Agent Trust Hub on Apr 29, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill requires the installation of the @membranehq/cli npm package, which is the official command-line interface for the Membrane platform.
  • [COMMAND_EXECUTION]: The skill utilizes the membrane CLI to perform authentication, manage connections, and execute actions. These are standard operations for interacting with the Membrane ecosystem.
  • [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection as it processes user-provided intent and external data payloads.
  • Ingestion points: External data is ingested through the --intent parameter in the action list command and the --input parameter in the action run command as seen in SKILL.md.
  • Boundary markers: No specific delimiters or boundary warnings are provided to separate instructions from untrusted data in command arguments.
  • Capability inventory: The skill allows for command execution via the CLI and the ability to trigger arbitrary actions on the Membrane platform.
  • Sanitization: There is no documentation or evidence of data sanitization before interpolation into shell commands.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 29, 2026, 06:36 AM