kelly-invoice-sheet

Pass

Audited by Gen Agent Trust Hub on Aug 25, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes Node.js scripts (import_batch.mjs, export_decisions.mjs) to perform automated data import and export tasks. These scripts are invoked via the CLI and are restricted to specific operations: reading local JSON batches, writing to the Busabase API, and generating CSV/JSON exports in a local directory. This aligns with the skill's primary function.
  • [DATA_EXFILTRATION]: Network activity is restricted to communication with the Busabase platform. The skill does not send invoice data or local files to arbitrary external servers. Credential handling for the Busabase API (BUSABASE_API_KEY, BUSABASE_SPACE_ID) follows best practices by relying on the execution environment rather than static files.
  • [EXTERNAL_DOWNLOADS]: Dependencies are restricted to well-known, versioned packages from the official npm registry, such as hono and busabase-sdk. There are no instances of unverified script downloads or piped remote shell execution (e.g., curl | bash).
  • [INDIRECT_PROMPT_INJECTION]: While the skill ingests external data from invoices and receipts, it mitigates injection risks by using structured JSON schemas and a dedicated validation engine (validateInvoicesShape). This ensures that extracted data is treated as values within a rigid table structure rather than executable instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 25, 2026, 10:35 AM
Security Audit — agent-trust-hub — kelly-invoice-sheet