chengfeng-export

Pass

Audited by Gen Agent Trust Hub on Aug 6, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill orchestrates video rendering and verification by executing vendor-provided Node.js scripts (videocut-cli.cjs, ensure-running.cjs) and system utilities (ffmpeg, ffprobe). It also performs file system operations such as creating hard links using ln on macOS or New-Item on Windows, and updating project configuration files based on file fingerprints.
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it ingests and processes various project configuration files to determine its execution path.
  • Ingestion points: The skill reads project.json, workbench.json, edit-list.json, subtitles.json, and visuals.json from the project directory.
  • Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the provided skill text to isolate data from instructions.
  • Capability inventory: The agent possesses capabilities for shell execution (node, ffmpeg, ffprobe) and file system modification.
  • Sanitization: The instructions do not specify any validation or sanitization logic for the data extracted from the ingested JSON files before processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 6, 2026, 01:08 AM
Security Audit — agent-trust-hub — chengfeng-export