scene-craft

Pass

Audited by Gen Agent Trust Hub on Nov 2, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes a local CLI tool (story) or a bundled Node.js fallback script (../story-maintenance/scripts/story.js) to perform project maintenance tasks like reindexing and word counts. The instructions demonstrate security awareness by mandating the use of the node binary directly instead of bun or package scripts to prevent the unintended loading of environment files or configuration-based code execution.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external content such as reader notes, editor comments, and imported text. This is a common attack surface for indirect prompt injection; however, the instructions explicitly command the agent to treat such content as data to be weighed and never as authoritative instructions, mitigating the risk of accidental obedience.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 2, 2026, 04:52 AM