univer-integrate

Pass

Audited by Gen Agent Trust Hub on Aug 17, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill includes instructions and templates that fetch official @univerjs library components from the NPM registry and load UMD bundles from unpkg.com, a well-known and trusted content delivery network.
  • [COMMAND_EXECUTION]: Integration templates and setup guides use standard package management and execution commands such as npm install, vite, and tsx to manage project dependencies and run application logic.
  • [PROMPT_INJECTION]: The skill provides mechanisms for processing external workbook and document data. 1. Ingestion points: univerAPI.createWorkbook() and CSV import processes. 2. Boundary markers: The documentation recommends validating untrusted JSON snapshots against a project-specific schema. 3. Capability inventory: Includes workbook manipulation, formula calculation, and network operations via optional plugins. 4. Sanitization: The skill explicitly advises users to validate untrusted JSON and escape potentially dangerous formula characters in CSV data before processing or exporting.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 17, 2026, 03:35 AM
Security Audit — agent-trust-hub — univer-integrate