movie-director
Pass
Audited by Gen Agent Trust Hub on Aug 14, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the
renkuCLI extensively to manage movie production workflows. This includes inspecting project state (e.g.,renku project current), managing casting and location data (e.g.,renku cast list), and coordinating media generation tasks. The commands are scoped to the project's own management utility and are used as intended for the skill's primary purpose. - [SAFE]: The instructions include specific safety controls for network operations, requiring the agent to request explicit permission before communicating with the local Studio server at
http://localhost:5173. It also correctly identifies thatlocalhostaccess is considered a network operation within the sandbox environment. - [SAFE]: The skill implements a strict approval gate for media production, forbidding the execution of paid generation tasks without first obtaining a cost estimate and explicit user confirmation. This prevents unintended costs and ensures user oversight of AI-generated content.
- [SAFE]: The instructions emphasize data integrity by requiring the agent to verify task completion by reading back the project state through the CLI rather than relying on natural language summaries from other skills.
Audit Metadata