short-drama-image-prompts

Pass

Audited by Gen Agent Trust Hub on Aug 19, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill's operational logic is entirely local. It generates Markdown and JSONL specifications for image prompts without initiating network connections or interacting with external APIs.
  • [SAFE]: The included Python validation script (image_prompt_check.py) uses only Python standard library modules. It implements a security feature that actively scans input records for 'provider execution fields' (such as api_key, access_token, secret) to prevent accidental credential leakage in prompt specifications.
  • [SAFE]: There are no signs of prompt injection, obfuscation, or persistence mechanisms. The skill follows a '职责分离' (separation of duties) principle, explicitly stating that it does not handle image production or model invocation itself.
  • [SAFE]: The instructions and scripts focus on 'reviewed invariants' and 'structural invariants', which are best practices for data integrity in automated workflows.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 19, 2026, 12:21 PM
Security Audit — agent-trust-hub — short-drama-image-prompts