comfy-custom-nodes-validation
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill outlines a process for validating and executing generated code, which presents a surface for indirect prompt injection.
- Ingestion points: The agent processes custom node files as described in SKILL.md.
- Boundary markers: There are no instructions to treat the code as untrusted data or to ignore embedded prompts.
- Capability inventory: The process involves importing Python code and running workflows, which are powerful actions.
- Sanitization: The skill does not specify sanitization or isolation for the code being tested.
Audit Metadata