keynote-translate

Pass

Audited by Gen Agent Trust Hub on Sep 1, 2026

Risk Level: SAFE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests text content from user-provided Keynote (.key) files for translation. While this presents a surface for potential indirect prompt injection (where instructions hidden in presentation text could influence the model), the workflow is tightly scoped to translation tasks. The use of AppleScript to re-insert text into the Keynote application further isolates the output from sensitive system functions.
  • [COMMAND_EXECUTION]: The skill uses AppleScript to automate Apple Keynote. This interaction is limited to the Keynote application's object model (slides, text items, shapes, and styles). The scripts provided are standard for macOS application automation and do not contain calls to execute arbitrary shell commands or modify system settings.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 1, 2026, 03:51 AM
Security Audit — agent-trust-hub — keynote-translate