orwell-perspective

Pass

Audited by Gen Agent Trust Hub on Aug 28, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill instructions and examples are focused entirely on linguistic analysis and writing improvement. No malicious patterns, command execution, or unauthorized data access behaviors were detected.
  • [EXTERNAL_DOWNLOADS]: The skill specifies an installation method npx skills add nuwa-skills/orwell-skill. This refers to the vendor's own package repository and represents standard installation for this author's skills.
  • [INDIRECT_PROMPT_INJECTION]: The skill contains a surface for processing untrusted data.
  • Ingestion points: Processes user-provided text for review and improvement in SKILL.md.
  • Boundary markers: Includes instructions to only trigger the perspective when explicitly requested by the user.
  • Capability inventory: The skill possesses no capabilities for command execution, file system modification, or network access.
  • Sanitization: None present, but the lack of exploitable tools renders the risk safe.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 28, 2026, 11:12 AM
Security Audit — agent-trust-hub — orwell-perspective