longform-highlight-select

Pass

Audited by Gen Agent Trust Hub on Jun 16, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill's entrypoint and examples demonstrate the execution of a local TypeScript harness script (scripts/harness/longform-highlight-select.ts) using node and tsx. This is the intended operation for processing the input data.
  • [SAFE]: The skill operates entirely on local files, processing word-level timestamps and media analysis data to produce highlight candidates. No indicators of network exfiltration, obfuscation, or safety guideline violations were found.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 16, 2026, 08:11 AM
Security Audit — agent-trust-hub — longform-highlight-select