capcut-export

Pass

Audited by Gen Agent Trust Hub on Sep 20, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTIONPERSISTENCE
Full Analysis
  • [COMMAND_EXECUTION]: The script scripts/export-capcut-template.py uses subprocess.run to call ffmpeg for media processing tasks, such as splitting video files into card-specific clips and generating a cover thumbnail. It also executes pgrep to verify that the CapCut application is closed before modifying project files.
  • [EXTERNAL_DOWNLOADS]: The skill requires the user to install the pyJianYingDraft and pillow Python packages from public registries to enable draft creation and image manipulation functionality.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external data that could potentially contain malicious instructions.
  • Ingestion points: Untrusted text data enters the context via the transcript.json file processed in scripts/export-capcut-template.py.
  • Boundary markers: The script does not use specific markers to delimit or ignore instructions embedded within the transcript text.
  • Capability inventory: The script has the capability to write to the file system and execute shell commands via subprocess.run.
  • Sanitization: The text processing logic in build_captions does not include sanitization or escaping of the input text.
  • [PERSISTENCE]: The script modifies the root_meta_info.json file (the CapCut project registry) to ensure new drafts appear in the application UI. The script performs a backup of this file before any changes are made.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 20, 2026, 08:46 AM