skills/inf-sh/skills/nano-banana/Gen Agent Trust Hub

nano-banana

Pass

Audited by Gen Agent Trust Hub on Feb 27, 2026

Risk Level: SAFEREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The skill downloads and executes an installation script from https://cli.inference.sh using a shell pipe. This is the vendor's official distribution method for the infsh CLI tool and is considered safe as it originates from the skill author's infrastructure.
  • [EXTERNAL_DOWNLOADS]: Fetches necessary binaries and configuration from dist.inference.sh and cloud.inference.sh. It also references the installation of related skills via npx from the inference-sh organization.
  • [COMMAND_EXECUTION]: Utilizes the infsh CLI to run AI applications and manage authentication via infsh login.
  • [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface. Ingestion points: The images parameter accepts external URLs in SKILL.md. Boundary markers: No explicit delimiters or ignore-instructions are used for image data. Capability inventory: Bash command execution via infsh. Sanitization: No content or metadata validation is performed on the input images before processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 27, 2026, 06:13 PM