webnovel-writing

Pass

Audited by Gen Agent Trust Hub on Jun 4, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill includes instructions to use a local Python script (scripts/search_corpus_examples.py) to search the novel database. This is a standard feature for retrieving relevant writing examples from the local environment and is scoped to the provided dataset.
  • [EXTERNAL_DOWNLOADS]: The skill documentation references external platforms such as socialistic.ai for skill sharing and star-history.com for analytics. These are legitimate services and do not involve untrusted code execution.
  • [INDIRECT_PROMPT_INJECTION]: The skill architecture involves reading and analyzing fictional content from a provided dataset of web novel articles. While this content originates from external platforms (e.g., Zhihu), it consists of standard fictional stories and benign metadata (Record Numbers), posing no security risk to the agent.
  • [SAFE]: Minor content obfuscation (e.g., using pinyin like 'sha' instead of the character '杀') is present in some reference materials. This is a standard practice in the Chinese web novel community to navigate platform filters and does not constitute a security concern for the agent's logic.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 4, 2026, 03:13 PM
Security Audit — agent-trust-hub — webnovel-writing