render-crucible-media

Pass

Audited by Gen Agent Trust Hub on Nov 19, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references the installation of the @codepawlhq/crucible-denoise-oidn NPM package and the execution of a script to fetch OIDN binaries. These are official vendor resources provided for render denoising.
  • [COMMAND_EXECUTION]: Technical documentation provides examples for using scripts/render-media.mjs and ffprobe. These are intended for developer use within the project repository to verify render outputs.
  • [INDIRECT_PROMPT_INJECTION]: The skill operates by ingesting scene recipes and session data. While this constitutes an attack surface for untrusted data, the behavior is core to the skill's functionality and no exploitable instructions were identified.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 19, 2026, 10:49 AM