basin

Pass

Audited by Gen Agent Trust Hub on Apr 22, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Fetches and installs the Membrane CLI (@membranehq/cli) from the npm registry. This is a known tool from the vendor used for authentication and API management.
  • [COMMAND_EXECUTION]: Utilizes the 'membrane' CLI to perform operations such as authentication, action discovery, and executing Basin API requests.
  • [PROMPT_INJECTION]: Vulnerable to indirect prompt injection from processing external data.
  • Ingestion points: External form submission data processed in SKILL.md via 'list-submissions' or 'get-submission'.
  • Boundary markers: Absent; the skill lacks delimiters or instructions to ignore embedded instructions in the form data.
  • Capability inventory: The skill can manage forms and webhooks or perform arbitrary requests through 'membrane request'.
  • Sanitization: Absent; no validation or escaping of external data is provided.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 22, 2026, 06:51 AM