mobile-kotlin-multiplatform
Pass
Audited by Gen Agent Trust Hub on Jul 8, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The 'Agent Protocol' section contains instructions to omit preambles, postambles, and explanations, and to 'compress output.' While likely intended for efficiency, these are behavior overrides that restrict the agent's natural explanatory functions and can be categorized as minor prompt injections.
- [SAFE]: The technical instructions for Kotlin Multiplatform, including the use of Ktor, SQLDelight, and Compose Multiplatform, follow industry best practices and do not contain malicious code.
- [SAFE]: No evidence of data exposure, credential harvesting, or unauthorized network operations was found within the skill or its associated references.
- [SAFE]: A large volume of repetitive technical filler was found in eight reference files (e.g., architecture-patterns.md). These sections repeat generic technical jargon and unrelated code snippets (Go, Rust, Python) but were determined to be benign technical noise rather than a concealment or poisoning vector.
Audit Metadata