scriptwriter-skill

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to maintain series continuity by reading existing episode files and processing user-supplied story concepts.
  • Ingestion points: The workflow involves using ls script/ to identify previous works and then reading those files (script/ep*-*.md). User input provides the core themes and plot directions for new content.
  • Boundary markers: The instructions lack specific delimiters or "ignore embedded instructions" warnings for the content read from files.
  • Capability inventory: The skill uses Read, Write, and Bash (restricted to ls) tools to manage the script files.
  • Sanitization: No explicit sanitization or validation of the ingested script content or user input is described.
  • [COMMAND_EXECUTION]: The skill instructs the agent to use the Bash tool to execute ls script/ as part of the continuity check process. This command is restricted to listing files within the project's script directory and does not involve executing external or untrusted code.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 02:56 PM