scrivener-scriptwriting
Pass
Audited by Gen Agent Trust Hub on Jun 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill invokes a local Python CLI tool specifically for reading Scrivener project files. This is an intended functionality of the skill and is constrained to local operations.
- [PROMPT_INJECTION]: The skill is designed to ingest and process text from external Scrivener projects, creating a surface for indirect prompt injection. However, no evidence of malicious intent was found.
- Ingestion points: Scrivener project files (.scriv) read via the
readcommand as mentioned inSKILL.mdandreferences/scriptwriting.md. - Boundary markers: Not explicitly defined in the current instructions.
- Capability inventory: Limited to reading file content and metadata; other modifying commands are currently listed as planned/unavailable.
- Sanitization: No explicit sanitization or filtering of project content is mentioned in the provided skill files.
Audit Metadata