quark-torch-model-intake
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill serves as a metadata descriptor and pointer to implementation logic for analyzing machine learning models. No malicious code, prompt injections, or unauthorized data access patterns were detected.- [INDIRECT_PROMPT_INJECTION]: The skill's primary function is to intake and analyze external model files, which constitutes a standard data ingestion surface. 1. Ingestion points: HuggingFace model files (safetensors, transformers) as specified in the description. 2. Boundary markers: None defined in this manifest. 3. Capability inventory: Architecture detection and metadata preparation; no active execution tools are defined in this manifest. 4. Sanitization: Not specified in this file.
Audit Metadata