agnes-ai-generation
Pass
Audited by Gen Agent Trust Hub on Jun 22, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The script
scripts/agnes_api.pyperforms network requests toapihub.agnes-ai.com. These operations are transparent and necessary for the skill to communicate with the Agnes AI service as intended. - [COMMAND_EXECUTION]: The skill provides a standalone Python script (
scripts/agnes_api.py) for the agent to execute. This script facilitates API interactions, manages prompt translation, and validates input parameters such as image dimensions and video frame rates. - [PROMPT_INJECTION]: The skill processes user-supplied prompts which are then transmitted to external AI models, presenting a theoretical surface for indirect prompt injection.
- Ingestion points: User input is received via the
--promptargument inscripts/agnes_api.py. - Boundary markers: The script does not implement specific delimiters to separate user content within the API request payload.
- Capability inventory: The skill possesses network access via Python's standard
urlliblibrary. - Sanitization: Prompts are transmitted as JSON-encoded strings, providing standard protection against basic payload confusion.
Audit Metadata