short-drama-review-normalizer

Pass

Audited by Gen Agent Trust Hub on Aug 7, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No security issues detected. The skill functions as a deterministic Python-based consent validator for internal workflows.
  • [COMMAND_EXECUTION]: The skill executes a local Python script defined as its entry point. This script uses only Python standard libraries and performs no subprocess calls, file system modifications, or shell executions.
  • [DATA_EXFILTRATION]: The skill has disable-model-invocation: true and contains no network-capable code (e.g., requests, urllib). It processes input from standard input and returns a classification block to standard output, maintaining a closed execution environment.
  • [PROMPT_INJECTION]: The skill is a defense against prompt injection in larger pipelines. It replaces non-deterministic LLM consent checks with strict regex-based matching to ensure that external media generation only occurs after explicit, unambiguous user approval.
  • [EXTERNAL_DOWNLOADS]: The skill does not define any external dependencies in its configuration and does not perform any runtime downloads.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 7, 2026, 08:35 PM
Security Audit — agent-trust-hub — short-drama-review-normalizer