mindspeed-fsdp2-model-migration
Pass
Audited by Gen Agent Trust Hub on Jun 15, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is designed to process external model source code, creating an indirect prompt injection surface.
- Ingestion points: The skill reads source model files and asset paths provided as input (e.g.,
source_model_file,runtime_assets.model_path). - Capability inventory: The agent is instructed to create model plugins and write new Python source files to the file system.
- Boundary markers: The instructions do not define clear boundaries or markers to isolate external source code from the agent's instructions, increasing the risk of the agent following embedded malicious directives.
- Sanitization: There is no mention of sanitizing or validating the contents of the source code before it is processed by the agent.
Audit Metadata