xero

Pass

Audited by Gen Agent Trust Hub on Mar 11, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the membrane CLI tool to execute actions and proxy requests to the Xero API, which is the intended core functionality of the integration.\n- [EXTERNAL_DOWNLOADS]: The skill recommends installing the @membranehq/cli package from the npm registry. This is a verified resource belonging to the skill's authoring organization.\n- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes data from external Xero records (such as invoices, contacts, and transactions) which could contain adversarial instructions.\n
  • Ingestion points: Data retrieved via actions like list-invoices, list-contacts, and get-invoice in SKILL.md.\n
  • Boundary markers: No delimiters or explicit instructions are present to distinguish between external data and system commands.\n
  • Capability inventory: The skill can execute actions via membrane action run and perform network requests via membrane request across all integration logic.\n
  • Sanitization: There is no evidence of data sanitization, validation, or escaping logic to prevent malicious content within Xero data from influencing the agent's behavior.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 11, 2026, 09:14 AM