paperzilla

Pass

Audited by Gen Agent Trust Hub on Aug 17, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references installation paths for the pz CLI tool through official GitHub repositories (paperzilla-ai/pz) and standard package managers like Homebrew and Scoop.
  • [COMMAND_EXECUTION]: Documents the use of the pz CLI tool to perform actions such as listing projects, retrieving paper summaries, and managing research feeds.
  • [PROMPT_INJECTION]: The skill ingests external research content, which constitutes a standard surface for indirect prompt injection. (1) Ingestion points: Research papers and project feeds via pz paper and pz feed (SKILL.md). (2) Boundary markers: None specified. (3) Capability inventory: pz CLI commands. (4) Sanitization: No explicit sanitization of external paper content is mentioned.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 17, 2026, 02:13 AM