wechat-sticker-production

Pass

Audited by Gen Agent Trust Hub on Oct 19, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses subprocess.run to orchestrate its workflow, calling internal scripts (e.g., compact_wechat_sticker.py) and standard media utilities like ffmpeg. These calls use argument lists rather than shell strings, preventing shell injection vulnerabilities.
  • [EXTERNAL_DOWNLOADS]: The skill interacts with a local H3 API for video rendering and utilizes the Chrome DevTools Protocol (CDP) to automate interactions with the official WeChat Sticker Platform. It explicitly restricts its own rendering API calls to localhost to prevent unauthorized network access.
  • [INDIRECT_PROMPT_INJECTION]: While the skill processes user-supplied labels and image references, it implements significant technical constraints and validation checks. It audits dimensions, byte sizes, and loop integrity of artifacts before use, and uses structured JSON for layout properties to maintain strict data boundaries.
  • [SAFE]: The skill demonstrates strong security hygiene by implementing resource gates (RAM and swap monitoring), utilizing file-level locking for shared GPU access, and providing clear instructions to exclude sensitive account credentials and screenshots from version control.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 19, 2026, 06:27 PM