matterport

Pass

Audited by Gen Agent Trust Hub on Apr 22, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the @membranehq/cli to manage connections and execute API actions. This is the vendor's official command-line interface and is used here for its intended purpose.\n- [EXTERNAL_DOWNLOADS]: The instructions guide the installation of the @membranehq/cli package from the NPM registry. This is a trusted dependency provided by the skill's author.\n- [PROMPT_INJECTION]: The skill processes data from external Matterport Spaces, which introduces a surface for indirect prompt injection.\n
  • Ingestion points: Data retrieved through the membrane action run and membrane request commands in the SKILL.md file.\n
  • Boundary markers: No explicit delimiters are used to wrap or isolate external Matterport content in the agent prompt.\n
  • Capability inventory: The skill uses the membrane CLI to perform network operations and execute pre-defined actions on the Matterport platform.\n
  • Sanitization: There is no documented sanitization or validation logic to filter potentially malicious instructions within Matterport data before it is processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 22, 2026, 01:42 AM