cml-lab-lifecycle
Pass
Audited by Gen Agent Trust Hub on Mar 18, 2026
Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [PROMPT_INJECTION]: Indirect Prompt Injection Surface.
- Ingestion points: The
import_labtool accepts and processes external topology data in YAML format. - Boundary markers: The skill does not include specific instructions to the agent to ignore or isolate instructions that might be embedded within the YAML topology data.
- Capability inventory: The skill has extensive control over the CML environment, including the ability to create, start, stop, and delete labs, and export node configurations.
- Sanitization: There is no documentation regarding the validation or sanitization of the YAML input before it is processed by the CML server.
- [EXTERNAL_DOWNLOADS]: The skill requires the
cml-mcppackage to be installed via pip. This is a vendor-owned component for CML integration.
Audit Metadata