autoverse-cli

Pass

Audited by Gen Agent Trust Hub on Sep 16, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill facilitates the execution of the autoverse command-line tool for tasks such as academic paper searching, metadata resolution, and account management.
  • [EXTERNAL_DOWNLOADS]: The skill provides instructions for installing the autoverse package from an external repository using pipx.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes academic metadata (e.g., paper titles and abstracts) retrieved from external sources through the CLI.
  • Ingestion points: Data returned from the search, resolve, authors, and related commands in SKILL.md and references/commands.md.
  • Boundary markers: The agent is explicitly instructed to use the --json flag to ensure output is received in a structured format, separating data from control flow.
  • Capability inventory: Shell execution of the autoverse CLI and local file export functionality.
  • Sanitization: No specific content sanitization is described, though the reliance on JSON output mitigates basic injection risks.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 16, 2026, 02:26 AM