add-gemini-nano

Pass

Audited by Gen Agent Trust Hub on Aug 5, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs users to install the '@1mt/expo-on-device-ai' package via pnpm. This dependency is a specialized library for managing on-device LLM inference in Expo projects and is directly relevant to the skill's purpose of enabling Gemini Nano support.- [PROMPT_INJECTION]: The skill provides code templates that interpolate data into LLM prompts (e.g., 'Summarize: ...'). This represents a surface for indirect prompt injection. * Ingestion points: Data enters the model via the 'prompt' field in the 'generateText' function described in SKILL.md. * Boundary markers: No delimiters or instruction-isolation markers are included in the provided code examples. * Capability inventory: The skill facilitates text generation using on-device hardware. * Sanitization: No input sanitization or validation routines are demonstrated in the instruction text.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 5, 2026, 11:17 PM
Security Audit — agent-trust-hub — add-gemini-nano