deepvista-recipe-research-to-vistabook

Pass

Audited by Gen Agent Trust Hub on Mar 31, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill body consists of instructional documentation (a 'recipe') for using local tools. No malicious behavior or direct security threats were identified during the analysis.
  • [COMMAND_EXECUTION]: The skill utilizes the 'deepvista' CLI tool to interact with a knowledge base. The commands (search, get, run) are consistent with the tool's intended use-case and the author's identity (DeepVista-AI).
  • [PROMPT_INJECTION]: The workflow involves an indirect prompt injection surface where data retrieved from an external knowledge base is processed and used as input for subsequent tool execution.
  • Ingestion points: External card content is retrieved via 'deepvista vistabase get' as described in SKILL.md.
  • Boundary markers: Absent; the skill suggests summarizing findings into a plain context string without specific delimiters or isolation instructions.
  • Capability inventory: The skill calls 'deepvista vistabook +run', which executes secondary workflows based on the input.
  • Sanitization: No sanitization or validation of the retrieved content is mentioned before it is passed to the next stage.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 31, 2026, 03:32 AM