jianying-draft-effects

Pass

Audited by Gen Agent Trust Hub on Oct 26, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests external data from media catalogs and enums, which represents a potential attack surface for indirect prompt injection.
  • Ingestion points: Data enters the agent context via media.catalogue and media.enums as described in SKILL.md and references/workflow.md.
  • Boundary markers: The instructions lack explicit text delimiters for external data but mandate human verification ("确认点") and a "WorkflowPlan" for review before application.
  • Capability inventory: The skill utilizes media.official_resource_apply, timeline.effect, timeline.remove, and media.sfx to modify local video draft files.
  • Sanitization: The skill enforces strict parameter validation, stating that only directory-declared parameter names and ranges should be used, and requires normalization of inputs.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 26, 2026, 07:55 PM