drama-director
Pass
Audited by Gen Agent Trust Hub on Apr 25, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses local Python scripts (generate_image.py and generate_video.py) to interact with API endpoints and handle media generation tasks.\n- [EXTERNAL_DOWNLOADS]: The skill makes network calls to api.atlascloud.ai, the service provider's domain, to submit generation requests and poll for completion.\n- [PROMPT_INJECTION]: The skill processes untrusted user-provided material like novel passages to generate model instructions, presenting a surface for indirect prompt injection.\n
- Ingestion points: User story text is ingested via the agent in SKILL.md Step 1.\n
- Boundary markers: The skill lacks explicit delimiters to separate untrusted user data from the agent's logic instructions.\n
- Capability inventory: The skill has the ability to execute shell commands and perform network operations via local scripts.\n
- Sanitization: There is no evidence of validation or sanitization of user-provided content prior to its use in generative prompts.
Audit Metadata