qiuzhi2046-skill

Pass

Audited by Gen Agent Trust Hub on May 4, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is composed of descriptive markdown files that establish a persona and content strategy for an AI agent. No malicious patterns were identified during the audit.
  • [PROMPT_INJECTION]: Instructions are focused on maintaining a specific character voice and content structure. There are no attempts to override system prompts, bypass safety guidelines, or use adversarial role-play techniques.
  • [DATA_EXPOSURE]: The files contain no hardcoded credentials, environment variables, or sensitive file paths. Mentions of a Feishu wiki link to a public knowledge base used by the persona.
  • [REMOTE_CODE_EXECUTION]: No patterns for downloading and executing remote scripts were found. The 'One-Click' methodology described in the text refers to a content design philosophy (simplifying steps for users) rather than automated script execution.
  • [COMMAND_EXECUTION]: There are no shell commands, subprocess calls, or dynamic context injections present in the instructions or reference files.
  • [EXTERNAL_DOWNLOADS]: References to external repositories (GitHub) and documentation platforms (Feishu) are informational and do not trigger automated downloads or dependency installations.
Audit Metadata
Risk Level
SAFE
Analyzed
May 4, 2026, 01:54 PM
Security Audit — agent-trust-hub — qiuzhi2046-skill