editorial-package
Pass
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from various external sources which could contain instructions meant to influence the agent's behavior.\n
- Ingestion points: The workflow specifies reading external files such as the standalone contract, editorial package rules, edit plan, shot briefs, and screenplay dialogue (SKILL.md).\n
- Boundary markers: The instructions do not include boundary markers or explicit prompts for the agent to ignore instructions embedded within the data files.\n
- Capability inventory: The skill performs file system write operations, specifically creating
editorial_manifest.json,subtitles.srt, and updatingcue_sheet.csv(SKILL.md).\n - Sanitization: There are no instructions for sanitizing or validating the content of the ingested files before they are processed or written to the output manifest.
Audit Metadata