jianying-draft-export-prep

Pass

Audited by Gen Agent Trust Hub on Oct 26, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes a local CLI named jianying to perform video processing. This execution is strictly controlled via a mandatory capability handshake (version, doctor, capabilities) and requires explicit user authorization for native execution or external service calls.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests external project drafts and media paths. Potential risks are mitigated through structural validation (project verify), input probing, and a requirement to work within isolated file copies to prevent accidental data corruption or malicious overrides.
  • [SAFE]: The instructions explicitly prohibit dangerous behaviors, such as auto-installing software, generating Python scripts for draft manipulation, and collecting or recording user credentials.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 26, 2026, 07:56 PM