movie-director
Pass
Audited by Gen Agent Trust Hub on Oct 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external, untrusted data which presents an attack surface for indirect prompt injection. 1. Ingestion points: Supporting materials such as research files, notes, and documents are imported into the project context via the screenplay-supporting-material-importer. 2. Boundary markers: The instructions lack specific delimiters or warnings to ignore instructions that might be embedded within the imported source files. 3. Capability inventory: The skill and its specialists execute renku CLI commands and interact with external media providers (e.g., ElevenLabs). 4. Sanitization: No explicit sanitization or filtering of content within the imported supporting material is described in the coordinator workflow.
- [COMMAND_EXECUTION]: The skill uses the renku CLI to perform various project management and server operations. Evidence includes the use of commands like renku project current, renku director context, and renku studio stop to manage the production environment.
Audit Metadata