jianying-draft-edit

Pass

Audited by Gen Agent Trust Hub on Oct 26, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes shell commands via a local 'jianying' CLI tool to perform operations such as probing media files, running editing jobs, and verifying project structures. These commands (e.g., jianying job run, jianying media probe) are used to interact with the local video editing environment based on user input.
  • [INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect injection as it processes untrusted user data like file paths and media metadata.
  • Ingestion points: Untrusted data enters the context via user-provided absolute paths for media files and source drafts (SKILL.md).
  • Boundary markers: The instructions emphasize pre-execution probing and schema validation, though they do not specify explicit delimiters for the LLM to ignore embedded instructions in the processed data.
  • Capability inventory: The skill utilizes the 'jianying' CLI for file reading, writing to new paths, and project verification (workflow.md).
  • Sanitization: The workflow includes mandatory steps to validate input schemas, probe media facts using the CLI, and refuse unknown parameters or non-existent paths before any destructive or creative action (SKILL.md, Step 3).
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 26, 2026, 07:55 PM