vivadicta-search-and-rewrite

Pass

Audited by Gen Agent Trust Hub on Apr 20, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to use shell commands involving the vivadicta CLI and jq to search, retrieve, and manipulate local transcription data. These commands are standard for the tool's documented purpose.
  • [PROMPT_INJECTION]: The skill displays an indirect prompt injection surface because it processes untrusted transcription data through AI presets.
  • Ingestion points: Transcription content is retrieved from the user's history via vivadicta search and vivadicta get (SKILL.md).
  • Boundary markers: No delimiters or specific "ignore instructions" warnings are used when passing transcription text to the rewrite engine.
  • Capability inventory: The skill has the capability to read local transcription history and execute AI-based rewrites via the vivadicta CLI (SKILL.md).
  • Sanitization: No evidence of content sanitization or validation is provided in the instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 20, 2026, 04:09 PM