de-ai-writer

Pass

Audited by Gen Agent Trust Hub on Aug 28, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external text and interpolates it into an LLM prompt, which is a standard interface for this type of tool. * Ingestion points: The script scripts/deai.py reads data from files, command-line arguments, or standard input. * Boundary markers: Input text is delimited by a plain-text header 【用户文本】 in the final prompt, which provides limited isolation. * Capability inventory: The script can write to the local filesystem via the -o flag and make network requests to LLM API endpoints using urllib.request. * Sanitization: No specific filtering or escaping is applied to the user-supplied text before processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 28, 2026, 12:59 PM
Security Audit — agent-trust-hub — de-ai-writer