podcast-script-generator

Pass

Audited by Gen Agent Trust Hub on Jun 30, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill's primary function is text-to-text transformation. It provides detailed linguistic rules and constraints to ensure a specific 'broadcasting' tone, which are standard instructional guidelines and do not attempt to bypass AI safety filters.
  • [COMMAND_EXECUTION]: The skill involves reading local files using the agent's provided 'Read' tool. It suggests a default path within an Obsidian vault (/Users/ugreen/Documents/obsidian/每日播客/) for the user's convenience. This is a functional requirement for a document processing skill and does not constitute unauthorized access as it relies on user-provided or user-confirmed paths.
  • [DATA_EXFILTRATION]: There are no network operations, API calls, or patterns suggesting the unauthorized transmission of data to external servers. The skill operates entirely on the provided text and local reference files.
  • [REMOTE_CODE_EXECUTION]: No remote scripts, package installations, or dynamic code execution patterns (such as eval or shell execution) are present in the skill or its reference materials.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 30, 2026, 09:24 AM
Security Audit — agent-trust-hub — podcast-script-generator