abuselpdb

Pass

Audited by Gen Agent Trust Hub on Apr 29, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Downloads and installs the @membranehq/cli tool from the public npm registry to enable interaction with the Membrane platform.
  • [COMMAND_EXECUTION]: Uses the membrane command-line utility for various operations including user authentication, establishing service connections, and running integration actions.
  • [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection through its action discovery and creation mechanism:
  • Ingestion points: Natural language strings passed to membrane action list --intent and membrane action create (SKILL.md).
  • Boundary markers: None identified in the instructions to delimit user-provided intent from system commands.
  • Capability inventory: The skill can execute shell commands (membrane action run) and perform network operations via the Membrane platform.
  • Sanitization: No explicit sanitization or validation of the input strings is described.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 29, 2026, 01:44 AM
Security Audit — agent-trust-hub — abuselpdb